Abnormality detector, processing system, and method for detecting abnormalities

Abstract

To provide an abnormality detector, a processor, and a method for detecting abnormalities which allow an image to detect to be easily obtained.SOLUTION: An abnormality detector 40 includes: an imaging unit 36 for taking an image of far-infrared radiation emitted from a detection target including a part transparent to visible beams of light or far-infrared radiation reflected by the detection target; a determination unit 41C for determining the presence or absence of abnormalities in the appearances of the detection target on the basis of the result of imaging by the imaging unit 36; and an output unit 41E for outputting the result of determination by the determination unit 41C to the processor for performing predetermined processing to the detection target.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present disclosure relates to an abnormality detection device, a processing system including the abnormality detection device, and an abnormality detection method.

Background Art

[0002] An example of a substrate processing apparatus includes an imaging unit, a placement unit, an irradiation unit, and a monitoring unit. The substrate has transparency to visible light. The substrate is formed of, for example, glass. The substrate may have a transparent conductive film or a metal film. The placement unit places the substrate to be imaged within the imaging range. The irradiation unit irradiates the substrate placed within the imaging range with a laser beam, thereby generating at least one of reflected light and scattered light of the laser beam at an end of the substrate. The imaging unit images an image of the end of the substrate due to reflection or scattering of the laser beam. The monitoring unit determines whether or not a crack or the like has occurred in the substrate from the imaging result by monitoring the imaging result of the imaging unit (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the above-described substrate processing apparatus, since it is necessary to reflect or scatter the laser beam at the end of the substrate, the position of the irradiation unit with respect to the position of the substrate within the imaging range is restricted. In addition, high accuracy of the position of the irradiation unit with respect to the position of the substrate is required so that reflection or scattering of the laser beam can be caused. Therefore, an apparatus capable of more easily obtaining an image of the substrate is required. Note that such problems are common regardless of the shape of the imaging target.

Means for Solving the Problems

[0005] An abnormality detection device for solving the above problems includes an imaging unit that images far-infrared rays emitted by a detection target including a portion transparent to visible light or far-infrared rays reflected by the detection target, a determination unit that determines the presence or absence of an abnormality in the appearance of the detection target based on the imaging result of the imaging unit, and an output unit that outputs the determination result of the determination unit to a processing device that performs a predetermined process on the detection target.

[0006] A processing system for solving the above problems includes the above abnormality detection device and a processing device that performs a predetermined process on the detection target.

[0007] An abnormality detection method for solving the above problems includes imaging far-infrared rays emitted by a detection target including a portion transparent to visible light or far-infrared rays reflected by the detection target, determining the presence or absence of an abnormality in the appearance of the detection target based on the imaging result, and outputting the determination result to a processing device that performs a predetermined process on the detection target.

[0008] According to the above abnormality detection device, processing system, and abnormality detection method, since the far-infrared rays emitted or reflected by the detection target itself are imaged by the imaging unit, it is not necessary to illuminate the detection target to image it. As a result, it is not necessary to align the position of the irradiation unit with high accuracy with respect to the position of the detection target, so that an image of the detection target can be obtained more easily.

[0009] In the above abnormality detection device, the detection target may be a substrate having a flat plate shape, and the determination unit may determine the presence or absence of cracks in the contour of the substrate. According to the above abnormality detection device, since the imaging result of the imaging unit varies greatly depending on the presence or absence of cracks, the accuracy of the determination result by the determination unit is likely to be improved.

[0010] In the above-described abnormality detection device, the processing device includes a temperature control unit, and the temperature control unit is configured to control the temperature of the detection target or the surroundings of the detection target so that the temperature difference between the detection target and the surroundings of the detection target becomes larger than before the temperature control. The imaging unit may image the detection target while the detection target or the surroundings of the detection target are being temperature-controlled by the temperature control unit.

[0011] According to the above-described abnormality detection device, the boundary between the detection target and the surroundings of the detection target is emphasized by the temperature difference between the detection target and the surroundings of the detection target. Therefore, the difference between a detection target without an abnormality and a detection target with an abnormality is likely to be reflected in the imaging result of the imaging unit. As a result, the accuracy of the determination result by the determination unit is likely to be improved.

[0012] In the above-described abnormality detection device, the processing device includes a holding unit, and the holding unit is configured to hold only a part of the detection target. The imaging unit may image the detection target while the detection target is being held by the holding unit.

[0013] According to the above-described abnormality detection device, when the detection target has an abnormality in its appearance such that a part of the detection target is missing, the possibility that the missing part of the detection target detaches from the detection target is higher than when the holding unit holds the entire detection target. As a result, the abnormality in the appearance of the detection target is likely to be reflected in the imaging result. As a result, the accuracy of the determination result by the determination unit is likely to be improved.

Brief Description of the Drawings

[0014]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Embodiments for Carrying Out the Invention

[0015] With reference to FIGS. 1 to 7, an abnormality detection device, a processing system, and an abnormality detection method will be described.

[0016] [Processing Device] With reference to FIGS. 1 and 2, the processing device will be described. Hereinafter, a cluster tool type sputtering device, which is an example of the processing device, will be described. As shown in FIG. 1, a cluster tool type sputtering device 10 includes a transfer chamber 11, two sputtering chambers 12, a loading / unloading chamber 13, and two substrate processing chambers 14. The sputtering device 10 includes a first sputtering chamber 12A and a second sputtering chamber 12B as the two sputtering chambers 12.

[0017] Each of the processing chambers 12A, 12B, 13, and 14 is connected to the transfer chamber 11. A gate valve is located between the transfer chamber 11 and each of the processing chambers 12A, 12B, 13, and 14. When the gate valve opens, the transfer space defined by the transfer chamber 11 and the spaces defined by each of the processing chambers 12A, 12B, 13, and 14 communicate with each other. On the other hand, when the gate valve closes, the transfer space of the transfer chamber 11 and each of the processing chambers 12A, 12B, 13, and 14 are separated from each other.

[0018] The transfer chamber 11 is equipped with a transfer robot 11R. The transfer robot 11R is configured to be able to transfer substrates between each processing chamber connected to the transfer chamber 11 and the transfer chamber 11. The substrate is an example of a detection target. The transfer robot 11R transfers the substrate along a plane substantially parallel to the horizontal plane between the transfer chamber 11 and each processing chamber.

[0019] The loading / unloading chamber 13 is configured to be able to load a substrate before processing into the sputtering apparatus 10 from outside the sputtering apparatus 10 and unload a substrate after processing from inside the sputtering apparatus 10 to outside the sputtering apparatus 10. The loading / unloading chamber 13 is provided with an exhaust section 13E. When loading the substrate, with the loading / unloading chamber 13 separated from the transfer chamber 11, the loading / unloading port of the loading / unloading chamber 13 is opened, whereby the loading / unloading chamber 13 is opened to the atmosphere. Next, after the substrate before processing is loaded from the loading / unloading port, the loading / unloading port is closed. Subsequently, the inside of the loading / unloading chamber 13 is exhausted by the exhaust section 13E, whereby the inside of the loading / unloading chamber 13 is depressurized to approximately the same level as the transfer chamber 11. In contrast, when unloading the substrate, after the substrate after processing is transferred from the transfer chamber 11 to the loading / unloading chamber 13, the gate valve located between the loading / unloading chamber 13 and the transfer chamber 11 is closed. Next, after the inside of the loading / unloading chamber 13 is pressurized to atmospheric pressure, the loading / unloading port of the loading / unloading chamber 13 is opened. Then, the substrate after processing is unloaded from the loading / unloading port to outside the sputtering apparatus 10.

[0020] Each sputtering chamber 12A, 12B is provided with a cathode 20. Each sputtering chamber 12A, 12B is configured to be able to form a thin film on the surface of the substrate by sputtering treatment using the cathode 20. Each sputtering chamber 12A, 12B is provided with exhaust sections 12AE, 12BE for exhausting the processing space defined by the sputtering chambers 12A, 12B.

[0021] Each substrate processing chamber 14 is configured to be able to perform a predetermined process on a substrate. The process performed in the substrate processing chamber 14 may be a pre-process performed on the substrate before film formation in the sputtering chambers 12A and 12B, or a post-process performed on the substrate after film formation in the sputtering chambers 12A and 12B. The pre-process and the post-process may be, for example, a heat treatment or a cooling treatment. Each substrate processing chamber 14 includes an exhaust unit 14E that exhausts the processing space defined by the substrate processing chamber 14.

[0022] Referring to FIG. 2, the transfer chamber 11, the first sputtering chamber 12A, and the connection portion located between the transfer chamber 11 and the first sputtering chamber 12A will be described in more detail. In the present embodiment, the first sputtering chamber 12A is described as an example of a processing chamber provided with an abnormality detection device, but one or more of each of the sputtering chambers 12A and 12B and each substrate processing chamber 14 may be provided with an abnormality detection device.

[0023] As shown in FIG. 2, the transfer chamber 11 includes a vacuum chamber 11C that defines a transfer space 11S. In the transfer space 11S, the transfer robot 11R is disposed as described above. The transfer robot 11R transfers the substrate Sb before film formation from the transfer chamber 11 to the first sputtering chamber 12A, and transfers the substrate Sb after film formation from the first sputtering chamber 12A to the transfer chamber 11.

[0024] The first sputtering chamber 12A includes a vacuum chamber 12AC that defines a first film formation space 12AS. The vacuum chamber 12AC is an example of a processing chamber, and the first film formation space 12AS is an example of a processing space that houses a detection target. As described above, the first sputtering chamber 12A includes a cathode 20. The cathode 20 includes a target 21 and a backing plate 22. The target 21 is formed of, for example, a metal, a metal compound, or a mixture of a metal and a metal compound. The target 21 has a sputtered surface that is sputtered when forming a film on the substrate Sb. The backing plate 22 is joined to the target 21. The backing plate 22 has conductivity. The cathode 20 is attached to one surface of a side wall portion that defines the first film formation space 12AS. At least the sputtered surface of the target 21 of the cathode 20 is exposed to the first film formation space 12AS.

[0025] The thin film formed on the substrate Sb by sputtering of the target 21 may have transparency to visible light. When the thin film has transparency to visible light, the target 21 is formed of, for example, a transparent metal oxide. The thin film formed on the substrate Sb may reflect visible light. When the thin film reflects visible light, the target 21 is formed of, for example, a metal.

[0026] A target power supply 23 is connected to the backing plate 22. By applying a voltage to the backing plate 22 by the target power supply 23, a voltage is applied to the target 21 connected to the backing plate 22. The target power supply 23 may be, for example, a DC power supply or an AC power supply.

[0027] The first sputtering chamber 12A includes a support portion 31. The support portion 31 includes a support surface 31F for supporting the substrate Sb. The support surface 31F may be formed of a material different from the material forming the surface that contacts the support surface 31F on the substrate Sb. For example, the support surface 31F may be formed of aluminum oxide, and the substrate Sb may be a glass substrate. By forming the support surface 31F and the substrate Sb of different materials from each other, it is possible to increase the difference in heat capacity between the support surface 31F and the substrate Sb. For example, the support portion 31 has a flat plate shape, and the support surface 31F is one plane included in the support portion 31. When viewed from a viewpoint facing the support surface 31F, the support surface 31F supports the entire substrate Sb.

[0028] The support portion 31 includes a suction portion 32 and a temperature control portion 33. The suction portion 32 and the temperature control portion 33 are located inside the support portion 31. The suction portion 32 electrostatically sucks the substrate Sb placed on the support surface 31F to the support surface 31F. The suction portion 32 is an electrostatic chuck that sucks the substrate Sb using electrostatic force.

[0029] The temperature control portion 33 is configured to control the temperature of the substrate Sb or the periphery of the substrate Sb so that the difference between the temperature of the substrate Sb and the temperature of the periphery of the substrate Sb becomes larger before temperature control. In the present embodiment, the temperature control portion 33 is configured to be able to adjust the temperature of the support portion 31 including the support surface 31F. When the substrate Sb is placed on the support surface 31F, when viewed from a viewpoint facing the support surface 31F, the target region to be temperature-controlled by the temperature control portion 33 is larger than the area of the substrate Sb, and the substrate Sb is located within the target region. Therefore, the temperature control portion 33 can control the temperature of the periphery of the substrate Sb. The temperature control portion 33 can control the temperature of the region surrounding the entire circumference of the substrate Sb among the periphery of the substrate Sb when viewed from a viewpoint facing the support surface 31F. The temperature control portion 33 may be a cooling portion that lowers the temperature of the support portion 31 or a heating portion that raises the temperature of the support portion 31.

[0030] The first sputtering chamber 12A is provided with a position changing unit 34. The position changing unit 34 is configured to be able to change the position of the support portion 31 with respect to the cathode 20. The position changing unit 34 changes the position of the support portion 31 between a first position and a second position. When the support portion 31 is located at the first position, the support surface 31F is substantially orthogonal to the cathode 20. That is, when the support portion 31 is located at the first position, the support surface 31F is positioned substantially along the horizontal direction. In contrast, when the support portion 31 is located at the second position, the support surface 31F is substantially parallel to the cathode 20. That is, when the support portion 31 is located at the second position, the support surface 31F is positioned substantially along the vertical direction.

[0031] The first sputtering chamber 12A is provided with a lifting device 35. The lifting device 35 is configured to be able to change the position of the substrate Sb with respect to the support surface 31F of the support portion 31. The lifting device 35 changes the position of the substrate Sb between a third position and a fourth position. When the substrate Sb is located at the third position, the substrate Sb is in contact with the support surface 31F. On the other hand, when the substrate Sb is located at the fourth position, the substrate Sb is positioned upward by a predetermined distance from the support surface 31F.

[0032] The lifting device 35 includes a plurality of lifting pins 35A and a lifting mechanism 35B. Each lifting pin 35A has a tip portion that contacts the substrate Sb. By contacting the substrate Sb, each lifting pin 35A positions the substrate Sb above the support surface 31F and holds the substrate Sb in a state where the substrate Sb is located at the fourth position. The lifting pin 35A is an example of a holding portion configured to hold only a part of the substrate Sb. The lifting mechanism 35B is configured to be able to change the position of the tip portion of the lifting pin 35A with respect to the support surface 31F along the vertical direction.

[0033] When the substrate Sb before film formation is transferred from the transfer robot 11R to the support part 31, and when the substrate Sb after film formation is transferred from the support part 31 to the transfer robot 11R, the elevating mechanism 35B raises the elevating pin 35A. Thereby, by positioning the elevating pin 35A at the fourth position, the substrate Sb is supported by the elevating pin 35A at the fourth position. When the position of the substrate Sb is changed from the fourth position to the third position, the elevating mechanism 35B lowers the elevating pin 35A, whereby the tip of the elevating pin 35A is positioned at a position below the support surface 31F. As a result, the substrate Sb is supported by the support surface 31F.

[0034] An imaging window 12AC1 is located on the upper wall of the vacuum chamber 12AC. The imaging window 12AC1 is fitted into a hole penetrating the upper wall of the vacuum chamber 12AC along the vertical direction. The imaging window 12AC1 is formed of a material that transmits far-infrared rays. The material forming the imaging window 12AC1 is preferably formed of the same material as the material forming the lens included in the imaging unit 36. Thereby, it is suppressed that the light imaged by the imaging unit 36 is absorbed or reflected by the imaging window 12AC1. The material forming the imaging window 12AC1 may be, for example, germanium (Ge), silicon (Si), zinc sulfide (ZnS), zinc selenide (ZnSe), chalcogenide glass, or the like.

[0035] Outside the vacuum chamber 12AC and at a position overlapping the imaging window 12AC1, the imaging unit 36 is arranged. The imaging unit 36 is configured to be able to image the far-infrared rays radiated by the substrate Sb or the far-infrared rays reflected by the substrate Sb. The imaging unit 36 is a far-infrared camera. The imaging unit 36 constitutes an abnormality detection device 40 (see FIG. 3) provided in the sputtering apparatus 10.

[0036] The vacuum chamber 12AC further includes a supply port 12AC2. The first sputtering chamber 12A further includes a gas supply unit 37. The gas supply unit 37 is connected to the supply port 12AC2. The gas supply unit 37 is a mass flow controller. The gas supply unit 37 is connected to a gas cylinder located outside the sputtering apparatus 10. The gas supply unit 37 supplies a plasma generation gas into the first film formation space 12AS.

[0037] The vacuum chamber 12AC further includes an exhaust port 12AC3 to which an exhaust unit 12AE is connected. The exhaust unit 12AE includes, for example, a valve and a pump.

[0038] When film formation on the substrate Sb is performed, the inside of the first film formation space 12AS is adjusted to a predetermined pressure by the exhaust by the exhaust unit 12AE and the plasma generation gas introduced from the gas supply unit 37. Next, when the target power supply 23 applies a voltage to the target 21, plasma is generated from the plasma generation gas. Then, the target 21 is sputtered by the charged particles in the plasma. When the sputtered particles emitted from the target 21 reach the substrate Sb due to the sputtering of the target 21, a thin film is formed on the substrate Sb.

[0039] The sputtering apparatus 10 includes a connection part 10P that connects the transfer chamber 11 to the first sputtering chamber 12A. The connection part 10P communicates the transfer space 11S of the transfer chamber 11 with the first film formation space 12AS of the first sputtering chamber 12A. A gate valve 10G is located at the connection part 10P. Since the connection part 10P is opened when the gate valve 10G is opened, the first film formation space 12AS is communicated with the transfer space 11S by the connection part 10P. On the other hand, since the connection part 10P is blocked when the gate valve 10G is closed, the first film formation space 12AS is separated from the transfer space 11S.

[0040] [Sputtering System] FIG. 3 is a block diagram showing the electrical configuration of a sputtering system, which shows the electrical configuration related to the detection of an abnormality of the substrate Sb. The sputtering system includes a sputtering apparatus 10 that performs a film-forming process on the substrate Sb, and an abnormality detection device 40 (see FIG. 3) that detects the presence or absence of an abnormality in the appearance of the substrate Sb.

[0041] The detection target in the abnormality detection device 40 of the present embodiment is a substrate Sb having a flat plate shape. The determination unit 41C (see FIG. 3) of the abnormality detection device 40 determines the presence or absence of cracks in the contour of the substrate Sb. As a result, since the imaging result of the imaging unit 36 varies greatly depending on the presence or absence of cracks, the accuracy of the determination result by the determination unit 41C is likely to be improved.

[0042] In the present embodiment, visible light means a wavelength band of 380 nm or more and 780 nm or less. In the present embodiment, far-infrared light means a wavelength band of 3000 nm or more and 1 mm or less. The substrate Sb, which is an example of the detection target, has transmissivity with respect to visible light, while having shielding property with respect to far-infrared light. The substrate Sb has a first transmittance with respect to visible light and a second transmittance with respect to far-infrared light, and the second transmittance is lower than the first transmittance.

[0043] The second transmittance is preferably 1 / 2 or less of the first transmittance. The first transmittance is preferably, for example, 75% or more, and more preferably 80% or more. The second transmittance is preferably 35% or less, and more preferably 30% or less. Also, the transmittance of the substrate Sb is preferably 10% or less, and more preferably 5% or less, in the wavelength band of 5000 nm or more among far-infrared light.

[0044] The material forming the substrate Sb may be, for example, glass as described above. Alternatively, the material forming the substrate Sb may be a transparent material other than glass that has transmissivity with respect to visible light and shielding property with respect to far-infrared light. The transparent material may be, for example, an inorganic compound or an organic compound. The inorganic compound may be, for example, a metal compound. The organic compound may be, for example, a synthetic resin.

[0045] The imaging unit 36 can detect light in a wavelength band of 3000 nm or more, and preferably can also detect light in a wavelength band of 5000 nm or more. The imaging unit 36 is an infrared camera as described above. The imaging unit 36 may be a thermal camera.

[0046] As shown in FIG. 3, the abnormality detection device 40 includes an imaging unit 36 and a processing unit 41. The processing unit 41 is electrically connected to the imaging unit 36 and the control unit 10C included in the sputtering device 10. The processing unit 41 performs processing related to determining the presence or absence of an abnormality in the appearance of the substrate Sb using the imaging result captured by the imaging unit 36.

[0047] The processing unit 41 includes a learning unit 41A, a storage unit 41B, a determination unit 41C, a management unit 41D, an output unit 41E, and an input unit 41F.

[0048] The learning unit 41A performs machine learning using the image captured by the imaging unit 36 so as to output the presence or absence of an abnormality in the appearance of the substrate Sb. The learning unit 41A may perform machine learning using only the image of the substrate Sb having no abnormality in appearance, or may perform machine learning using the image of the substrate Sb having no abnormality in appearance and the image of the substrate Sb having an abnormality in appearance. The learning unit 41A generates an abnormality detection model as a learning result.

[0049] The storage unit 41B holds the abnormality detection model generated by the learning unit 41A. The storage unit 41B may hold only one abnormality detection model, or may hold two or more abnormality detection models. When the storage unit 41B holds two or more abnormality detection models, for example, it may hold a first abnormality detection model, a second abnormality detection model, and a third abnormality detection model.

[0050] The first abnormality detection model is a learning model generated by the learning unit 41A for the substrate Sb before film formation. The second abnormality detection model is a learning model generated by the learning unit 41A for the substrate Sb on which a thin film transparent to visible light is formed. The third abnormality detection model is a learning model generated by the learning unit 41A for the substrate Sb on which a thin film reflecting visible light and far-infrared light is formed.

[0051] The determination unit 41C uses the abnormality detection model to determine whether the substrate Sb has an appearance abnormality based on the imaging result of the imaging unit 36 imaging the substrate Sb. The determination result of the determination unit 41C may be, for example, only information indicating the presence or absence of an abnormality in the appearance of the substrate Sb. Alternatively, the determination result of the determination unit 41C may include information representing the degree of abnormality in each pixel included in the image used for the determination.

[0052] The management unit 41D manages the process for the learning unit 41A to generate the abnormality detection model, the process for the storage unit 41B to hold the abnormality detection model, and the process for the determination unit 41C to determine the presence or absence of an abnormality in the appearance of the substrate Sb. The management unit 41D has an abnormality detection program, and by executing the program including the abnormality detection program, causes the processing unit 41 to function as the learning unit 41A, the storage unit 41B, and the determination unit 41C. Further, the management unit 41D causes the imaging unit 36 to image the substrate Sb at a predetermined timing by executing the program including the abnormality detection program.

[0053] The output unit 41E outputs the determination result of the determination unit 41C to the sputtering apparatus 10 that performs a film formation process on the substrate Sb. The output unit 41E outputs the determination result of the determination unit 41C to the control unit 10C of the sputtering apparatus 10.

[0054] The input unit 41F inputs an image including the substrate Sb imaged by the imaging unit 36. The input unit 41F inputs a signal output by the control unit 10C. The signal output by the control unit 10C is, for example, a trigger signal for starting the abnormality detection process by the abnormality detection device 40.

[0055] The processing unit 41 includes electronic circuits such as a CPU and an MPU. The processing unit 41 includes a storage such as an SSD and an HDD. The processing unit 41 includes memories such as a ROM, a RAM, and a register memory. The processing unit 41 may include an integrated circuit such as an ASIC and an FPGA. All of the processes to be executed by the processing unit 41 may be executed by the software included in the processing unit 41, or may be executed by a combination of the integrated circuit and the software included in the processing unit 41.

[0056] According to the abnormality detection device 40 of the present disclosure, the imaging unit 36 images the far-infrared rays radiated or reflected by the substrate Sb itself to be detected. Therefore, it is possible to clarify the boundary between the substrate Sb and the periphery of the substrate Sb due to the temperature difference between the substrate Sb and the periphery of the substrate Sb. Therefore, it is not necessary to illuminate the substrate Sb in order to image the substrate Sb. As a result, since it is not necessary to align the position of the irradiation unit with high accuracy with respect to the position of the substrate Sb, it is possible to more easily obtain an image of the substrate Sb.

[0057] The sputtering apparatus 10 includes a control unit 10C. The control unit 10C is electrically connected to the processing unit 41 included in the abnormality detection device 40. After generating a trigger signal for starting the abnormality detection process of the substrate Sb in the abnormality detection device 40, the control unit 10C outputs the trigger signal to the processing unit 41 included in the abnormality detection device 40.

[0058] The control unit 10C includes, for example, a film formation processing program for performing a film formation process on the substrate Sb. By executing the film formation processing program, the control unit 10C outputs a trigger signal for starting the abnormality detection process to the abnormality detection device 40. The control unit 10C outputs a trigger signal specific to the film formation processing program to the abnormality detection device 40. Therefore, when the abnormality detection device 40 holds a plurality of abnormality detection models, the determination unit 41C can use the abnormality detection model corresponding to the input trigger signal.

[0059] The control unit 10C includes electronic circuits such as a CPU and an MPU. The control unit 10C includes storage such as an SSD and an HDD. The control unit 10C includes memories such as a ROM, a RAM, and a register memory. The control unit 10C may include integrated circuits such as an ASIC and an FPGA. All of the processes executed by the control unit 10C may be executed by the software included in the control unit 10C, or may be executed by a combination of the integrated circuit and software included in the control unit 10C.

[0060] The sputtering apparatus 10 includes a display unit 15. The display unit 15 is electrically connected to the control unit 10C. The display unit 15 inputs the determination result output by the control unit 10C, and thereby outputs, for example, an image obtained by imaging the substrate Sb. The display unit 15 may be, for example, various displays.

[0061] [Abnormality Detection Method] With reference to FIGS. 4 and 5, an abnormality detection method will be described. FIG. 4 is a flowchart for explaining a usage method of an abnormality detection apparatus 40 including an abnormality detection method.

[0062] As shown in FIG. 4, the usage method of the abnormality detection apparatus 40 includes an installation step (step S11), an image acquisition step (step S12), a learning step (step S13), a determination step (step S14), and a diagnosis step (step S15).

[0063] In the installation step, the abnormality detection apparatus 40 is set in the sputtering apparatus 10, which is an example of a processing apparatus. The installation step includes, for example, a step of installing the imaging unit 36 in the sputtering apparatus 10, a step of electrically connecting the imaging unit 36 to the processing unit 41, a step of electrically connecting the processing unit 41 to the control unit 10C, a step of installing a program for controlling the abnormality detection apparatus 40 in the control unit 10C, and the like.

[0064] In the image acquisition step, an image used for generating an abnormality detection model by the learning unit 41A is acquired. In the image acquisition step, for example, only an image of the substrate Sb having no abnormality in appearance may be acquired, or an image of the substrate Sb having no abnormality in appearance and an image of the substrate Sb having an abnormality in appearance may be acquired.

[0065] In the learning step, the learning unit 41A generates an abnormality detection model using the image acquired in the image acquisition step. The learning step may include a step of evaluating the performance of the abnormality detection model generated by the learning unit 41A.

[0066] In the determination step, it is determined whether the substrate Sb has an abnormality in appearance using an abnormality detection method. The abnormality detection method includes imaging far-infrared rays, determining the presence or absence of an abnormality, and outputting a determination result. By imaging far-infrared rays, far-infrared rays emitted by the substrate Sb including a portion transparent to visible light or far-infrared rays reflected by the substrate Sb are imaged. By determining the presence or absence of an abnormality, the presence or absence of an abnormality in the appearance of the substrate Sb is determined based on the imaging result. By outputting the determination result, the determination result is output to a processing device that performs a predetermined process on the substrate Sb. In the present embodiment, the determination result is output to the sputtering apparatus 10 that performs a film forming process on the substrate Sb.

[0067] When the substrate Sb is imaged using an imaging unit that detects visible light, a laser beam is irradiated to the end portion of the substrate Sb in order to clarify the boundary between the substrate Sb and the periphery of the substrate Sb. In this case, it is necessary to irradiate the end portion of the substrate Sb with a laser beam so that reflection or scattering of the laser beam occurs at the end portion of the substrate Sb. Therefore, high accuracy is required for the position of the irradiation unit with respect to the position of the end portion of the substrate Sb.

[0068] In addition, in order to increase the brightness of the entire circumference of the substrate Sb at once, it is necessary to arrange a plurality of irradiation units around the substrate Sb while maintaining high positional accuracy. On the other hand, it is possible to increase the brightness of the entire circumference of the substrate Sb by changing the position of the substrate Sb with respect to an irradiation unit in which the irradiation position of the laser beam is fixed with respect to the processing space. However, when the substrate Sb has cracks, changing the position of the substrate Sb may scatter fragments of the substrate Sb into the processing space.

[0069] In this regard, according to the abnormality detection device 40 of the present embodiment, since the far-infrared rays radiated by the substrate Sb or the far-infrared rays reflected by the substrate Sb are imaged, an irradiation unit for irradiating the substrate Sb with light in order to image the substrate Sb is unnecessary. Therefore, it is not necessary to install the irradiation unit with high positional accuracy or to install many irradiation units. Further, according to the abnormality detection device 40 of the present embodiment, scattering of fragments of the substrate Sb into the processing space can also be suppressed.

[0070] In the diagnosis step, the performance of the abnormality detection model held in the storage unit 41B is diagnosed. As time elapses since the start of operation of the sputtering system, the performance of the abnormality detection model may deteriorate. Factors that cause performance degradation include, for example, aging deterioration of the imaging unit 36, contamination and deterioration of the imaging window 12AC1, model optimization due to changes in the processing process used in the sputtering apparatus 10, and model optimization due to changes in the level required for products including the substrate Sb.

[0071] When diagnosing the performance of the anomaly detection model, for example, an image of a substrate Sb including an anomaly in appearance is used to cause the determination unit 41C to determine whether or not the substrate Sb in the image has an anomaly in appearance. Then, the user of the sputtering system determines whether or not the result of the determination by the determination unit 41C is correct. Also, for example, an image of a substrate Sb not including an anomaly in appearance is used to cause the determination unit 41C to determine whether or not the substrate Sb in the image has an anomaly in appearance. Then, the user of the sputtering system determines whether or not the result of the determination by the determination unit 41C is correct. From the viewpoint of improving the accuracy in diagnosing the performance of the anomaly detection model, it is preferable to use an image of a substrate Sb including an anomaly in appearance and an image of a substrate Sb not including an anomaly in appearance to determine whether or not the determination by the determination unit 41C for each image is correct.

[0072] As a method for determining whether the performance of the anomaly detection model has deteriorated, for example, the anomaly detection model is applied to a plurality of substrates having appearance anomalies and a plurality of substrates having no appearance anomalies, a confusion matrix is created, and based on the confusion matrix, the probability of meeting the required performance requirements and the probability of deterioration from the conventional performance are calculated based on statistical methods, and a method of determining whether adjustment of the anomaly detection model is necessary can also be adopted. Alternatively, a subjective method can be adopted in which a normal image used for learning the anomaly detection model and a recently obtained normal image are visually compared, and if any change is observed, it is determined that the performance of the anomaly detection model should have deteriorated. When the performance of the anomaly detection model is appropriate, the anomaly determination process in the determination step is continued using the same anomaly detection model.

[0073] When a decrease in the performance of the anomaly detection model is recognized and an aging deterioration of the imaging unit 36 is recognized, the anomaly detection model used for the anomaly detection process is updated by performing the process again from the installation step.

[0074] When a decrease in the performance of the abnormality detection model is recognized, and there is a change in the processing process used in the sputtering apparatus 10 and it is necessary to install a new program, the abnormality detection model used for the abnormality detection process is updated by performing the process again from the installation step. When a decrease in the performance of the abnormality detection model is recognized, and there is a change in the processing process used in the sputtering apparatus 10 but it is not necessary to install a new program, the abnormality detection model used for the abnormality detection process is updated by performing the process again from the image acquisition step.

[0075] When a decrease in the performance of the abnormality detection model is recognized and it is not accompanied by both the aging deterioration of the imaging unit 36 and the change in the processing process, or when it is not accompanied by both the contamination or deterioration of the imaging window 12AC1 and the change in the processing process, the abnormality detection model used for the abnormality detection process is updated by performing the process again from the learning step.

[0076] FIG. 5 is a sequence chart for explaining the abnormality detection method included in the determination step. As shown in FIG. 5, the control unit 10C outputs a trigger signal to the abnormality detection device 40 (step S21), and subsequently, the processing unit 41 of the abnormality detection device 40 inputs the output trigger signal (step S31). Thereby, the abnormality detection process in the abnormality detection device 40 is started.

[0077] In the abnormality detection device 40, first, the imaging unit 36 images the substrate Sb (step S32). The imaging unit 36 may image the substrate Sb, for example, while the substrate Sb or the periphery of the substrate Sb is temperature-controlled by the temperature control unit 33. The imaging unit 36 images the substrate Sb in a state where the substrate Sb is in contact with the support surface 31F of the support unit 31. In this case, due to the temperature difference between the substrate Sb and the periphery of the substrate Sb, the boundary between the substrate Sb and the periphery of the substrate Sb is emphasized. Therefore, the difference between the substrate Sb without abnormalities and the substrate Sb with abnormalities is likely to be reflected in the imaging result of the imaging unit 36. As a result, the accuracy of the determination result by the determination unit 41C is likely to be improved.

[0078] Alternatively, the imaging unit 36 may image the substrate Sb while the substrate Sb is held by the lifting pins 35A. In this case, the imaging unit 36 may image the substrate Sb when the lifting pins 35A place the substrate Sb at the fourth position. Alternatively, the imaging unit 36 may image the substrate Sb until the lifting pins 35A move the substrate Sb from the fourth position to the third position. Or alternatively, the imaging unit 36 may image the substrate Sb until the lifting pins 35A move the substrate Sb from the third position to the fourth position.

[0079] In this case, when the substrate Sb has an abnormality in its appearance such that a part of the substrate Sb is chipped, the chipped portion of the substrate Sb is more likely to detach from the substrate Sb than when the lifting pins 35A hold the entire substrate Sb. As a result, the abnormality in the appearance of the substrate Sb is likely to be reflected in the imaging result. As a result, the accuracy of the determination result by the determination unit 41C is likely to be improved.

[0080] Next, the determination unit 41C of the processing unit 41 determines the presence or absence of an abnormality in the appearance of the substrate Sb based on the imaging result of the imaging unit 36 (step S33). At this time, as described above, the determination unit 41C may generate only the presence or absence of cracks at the outer edge of the substrate Sb as the determination result, or may generate the degree of abnormality, which is the degree of abnormality in each pixel included in the image that is the imaging result, as the determination result.

[0081] Then, the abnormality detection device 40 outputs the determination result generated by the determination unit 41C to the control unit 10C of the sputtering device 10 (step S34). Subsequently, after the control unit 10C inputs the output determination result (step S22), it outputs the determination result to the display unit 15 (step S23).

[0082] The display unit 15 receives the determination result output by the control unit 10C (step S41). Subsequently, the display unit 15 displays the received determination result (step S42). The display unit 15 displays an image corresponding to the received determination result. The display unit 15 can also directly acquire and display the determination result and the image from the abnormality detection device 40 using web technology. As a result, the flow of FIG. 4 can be realized only by the display unit 15 without going through the control unit 10C, so the development burden of the control unit and the like can be reduced. The display unit 15 may display, for example, only characters or symbols indicating whether the substrate Sb has an abnormality in appearance. Alternatively, the display unit 15 may display, together with the image of the substrate Sb used for abnormality detection, an image indicating the degree of abnormality of each pixel included in the image. By the display unit 15 displaying the determination result, the user of the sputtering system can grasp the determination result.

[0083] Note that the abnormality detection process may be performed on one substrate Sb at two or more different timings. For example, for the same substrate Sb, the abnormality detection process may be performed at a first timing before film formation and a second timing after film formation. In this case, the control unit 10C may cause the abnormality detection device 40 to perform the abnormality detection process twice by outputting the trigger signal to the abnormality detection device 40 only once. Alternatively, the control unit 10C may separately output a first trigger signal for causing the abnormality detection device 40 to perform the first abnormality detection process and a second trigger signal for causing the abnormality detection device 40 to perform the second abnormality detection process. Alternatively, the same abnormality detection model may be applied at each timing, or different abnormality detection models may be used.

[0084] [Test Example] The test example will be described with reference to FIGS. 6 and 7. FIG. 6 shows an imaging result of a glass substrate Sb without cracks and an example of a determination result. FIG. 7 shows an imaging result of a glass substrate Sb with cracks and an example of a determination result.

[0085] Of the three images included in FIGS. 6 and 7 respectively, the first images PA1 and PB1 are images of the substrate Sb captured by the imaging unit 36. The first images PA1 and PB1 may be those obtained by trimming and enlarging the captured images by the imaging unit 36, or those subjected to image processing such as adjusting brightness and contrast to improve the determination accuracy. The second images PA2 and PB2 are determination results obtained by determination using the abnormality detection model, and are heat maps showing the degree of abnormality of each pixel included in the first images PA1 and PB1. The third images PA3 and PB3 are images obtained by superimposing the second images PA2 and PB2 on the first images PA1 and PB1.

[0086] As shown in FIGS. 6 and 7, according to the abnormality detection device 40, since the far-infrared rays radiated by the substrate Sb are imaged, it was confirmed that an image with a clear boundary between the substrate Sb and the support surface 31F can be obtained. Further, according to the abnormality detection device 40, it was confirmed that the crack SbC of the substrate Sb is displayed as an abnormal portion HMA in the heat map.

[0087] As described above, according to one embodiment of the abnormality detection device, the processing system, and the abnormality detection method, the following effects can be obtained. (1) Since the imaging unit 36 images the far-infrared rays radiated or reflected by the substrate Sb itself, it is not necessary to illuminate the substrate Sb to image the substrate Sb. As a result, it is not necessary to align the position of the irradiation unit with high accuracy with respect to the position of the substrate Sb, so that an image of the substrate Sb can be obtained more easily.

[0088] (2) When the substrate Sb has a flat plate shape, since the imaging result of the imaging unit 36 varies greatly depending on the presence or absence of cracks, the accuracy of the determination result by the determination unit 41C is likely to be improved.

[0089] (3) Due to the temperature difference between the substrate Sb and the periphery of the substrate Sb, the boundary between the substrate Sb and the periphery of the substrate Sb is emphasized. Therefore, the difference between the substrate Sb without abnormality and the substrate Sb with abnormality is likely to be reflected in the imaging result of the imaging unit 36. As a result, the accuracy of the determination result by the determination unit 41C is likely to be improved.

[0090] (4) When the lifting pin 35A holds the substrate Sb, if the substrate Sb has an abnormality in appearance such that a part of the substrate Sb is chipped, the possibility that the chipped part of the substrate Sb detaches from the substrate Sb is higher than when the holding part holds the entire substrate Sb. As a result, the abnormality in the appearance of the substrate Sb is likely to be reflected in the imaging result. Consequently, the accuracy of the determination result by the determination unit 41C is likely to be enhanced.

[0091] Note that the above-described embodiment can be implemented with the following modifications. [Imaging unit] · The abnormality detection device 40 may include two or more imaging units 36. In this case, the vacuum chamber 12AC may include the same number of imaging windows 12AC1 as the imaging units 36, and one imaging unit 36 may be arranged for one imaging window 12AC1. Alternatively, the vacuum chamber 12AC may include one imaging window 12AC1, and a plurality of imaging units 36 may be arranged for the imaging window 12AC1. Alternatively, the vacuum chamber 12AC may include two or more imaging windows 12AC1, and a plurality of imaging units 36 may be arranged for each imaging window 12AC1.

[0092] · The imaging unit 36 may be located inside the vacuum chamber 12AC. In this case, the vacuum chamber 12AC does not necessarily have to have the imaging window 12AC1.

[0093] · The imaging unit 36 may be arranged in a processing chamber other than the first sputtering chamber 12A among the processing chambers provided in the sputtering apparatus. In this case, it is possible to perform an abnormality detection process on the substrate Sb located in the processing chamber other than the first sputtering chamber 12A. Note that in this case, it is preferable that the storage unit 41B of the processing unit 41 holds an abnormality detection model for the abnormality detection process in each processing chamber.

[0094] [Temperature control unit] · While the temperature control unit 33 can adjust the temperature of the substrate Sb, it may be configured not to adjust the temperature around the substrate Sb. For example, when viewed from a perspective facing the support surface 31F, the area of the temperature control unit 33 may be smaller than the area of the substrate Sb, and the thermal conductivity between the temperature control unit 33 and the substrate Sb may be higher than the thermal conductivity between the temperature control unit 33 and the outside of the temperature control unit 33 when viewed from a perspective facing the support surface 31F.

[0095] [Detection target] · The appearance abnormality of the substrate Sb may not be an abnormality that causes a change in the contour of the substrate Sb. For example, the appearance abnormality of the substrate Sb may be a crack formed in the substrate Sb, a scratch formed on the outer surface of the substrate Sb, or the like.

[0096] · The detection target of the abnormality detection device 40 may not be an object having a flat plate shape like the substrate Sb. That is, the detection target may be an object having an arbitrary shape. Even when the detection target has an arbitrary shape and has a portion transparent to visible light, an image including the outer edge of the detection target can be generated by the imaging unit 36 that detects far-infrared rays radiated from the detection target.

Explanation of reference numerals

[0097] 10... Sputtering apparatus 12A... First sputtering chamber 12AC... Vacuum chamber 33... Temperature control unit 35... Lifting device 35A... Lifting pin 35B... Lifting mechanism 36... Imaging unit 40... Abnormality detection device 41... Processing unit 41C... Judgment unit 41E... Output unit Sb... Substrate

Claims

1. An imaging unit that images far-infrared rays emitted by a detection target including a portion transparent to visible light, or far-infrared rays reflected by the detection target, A determination unit that determines the presence or absence of an abnormality in the appearance of the detection target based on the imaging result of the imaging unit, An output unit that outputs the determination result of the determination unit to a processing device that performs a predetermined process on the detection target, and comprising An abnormality detection device.

2. The detection target is a substrate having a flat plate shape, The determination unit determines the presence or absence of cracks in the contour of the substrate The abnormality detection device according to claim 1.

3. The processing device includes a temperature control unit, The temperature control unit is configured to control the temperature of the detection target or the surroundings of the detection target so that the temperature difference between the detection target and the surroundings of the detection target becomes larger than before temperature control, The imaging unit images the detection target while the detection target or the surroundings of the detection target are being temperature-controlled by the temperature control unit The abnormality detection device according to claim 1.

4. The processing device includes a holding unit, The holding unit is configured to hold only a part of the detection target, The imaging unit images the detection target while the detection target is being held by the holding unit The abnormality detection device according to claim 1.

5. An abnormality detection device according to any one of claims 1 to 4, and A processing device that performs a predetermined process on the detection target, and comprising A processing system.

6. Imaging far-infrared rays emitted by a detection target including a portion transparent to visible light, or far-infrared rays reflected by the detection target, Determining the presence or absence of an abnormality in the appearance of the detection target based on the imaging result, and Outputting the determination result to a processing device that performs a predetermined process on the detection target, and comprising An abnormality detection method.

Images