Monitoring polishing pad texture in chemical mechanical polishing

Through the machine learning-based optical monitoring system to measure and adjust the surface roughness of the polishing pad, the problems of glazing and inhomogeneity of the polishing pad surface are solved, and the polishing efficiency and product quality are improved.

CN118977201APending Publication Date: 2024-11-19APPLIED MATERIALS INC
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN202411005153.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2019-06-14
Filing Date
2020-03-16
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

During the chemical mechanical polishing process, the surface of the polishing pad becomes glazed due to the accumulation of slurry by-products and materials, resulting in a decrease in polishing rate and an increase in substrate non-uniformity. The existing adjustment technology is difficult to effectively solve the surface roughness of the polishing pad. Problems of inhomogeneity and uneven wear.

Method used

Using an optical monitoring system based on machine learning, the image of the polishing pad surface is captured by the imager and input it into the trained neural network model to generate measurement results of the polishing pad surface roughness, and adjust the adjustment parameters according to the measurement results to realize the polishing pad Improved uniformity of surface texture.

Benefits of technology

Accurate and rapid measurement and adjustment of the surface roughness of the polishing pads, reduce inhomogeneity between wafers, avoid the risk of pollution caused by contact measurement, and improve polishing efficiency and product quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118977201A_ABST
    Figure CN118977201A_ABST
Patent Text Reader

Abstract

An apparatus for chemical mechanical polishing includes a platform having a surface to support a polishing pad; a carrier head holding the substrate against the polishing surface of the polishing pad; a pad adjuster that presses the polishing body against the polishing surface; an in-situ polishing pad monitoring system including an imager disposed above the platform to capture an image of the polishing pad; and a controller configured to receive the image from the monitoring system and generate a measurement of polishing pad surface roughness based on the image. The controller may generate a measurement of surface roughness using machine learning-based image processing.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] This application is a divisional application of a Chinese patent application (PCT application number PCT / US2020 / 022977) with an application date of March 16, 2020, application number 202080021537.X, and name “Monitoring the Polishing Pad Texture in Chemical Mechanical Polishing”. Technical Field

[0002] The present disclosure relates to optical monitoring of polishing pads used in chemical mechanical polishing. Background Art

[0003] Integrated circuits are typically formed on a substrate by sequentially depositing a conductive layer, a semiconductive layer, or an insulating layer on a silicon wafer. Various manufacturing processes require planarizing the layers on the substrate. For example, one manufacturing step involves depositing a conductive filler layer on a patterned insulating layer to fill the grooves or holes in the insulating layer. The filler layer is then polished until the raised pattern of the insulating layer is exposed. After planarization, the portion of the conductive filler layer that remains between the raised pattern of the insulating layer forms vias, plugs, and lines that provide conductive paths between thin film circuits on the substrate.

[0004] Chemical mechanical polishing (CMP) is an accepted planarization method. This planarization method generally requires mounting the substrate on a carrier head. The exposed surface of the substrate is placed against a rotating polishing pad. The carrier head provides a controllable load on the substrate to push the substrate against the polishing pad. A polishing liquid (such as a slurry with abrasive particles) is supplied to the surface of the polishing pad.

[0005] After performing a CMP process for a period of time, the surface of the polishing pad may become glazed due to accumulation of slurry byproducts and / or material removed from the substrate and / or polishing pad. Glazing may reduce the polishing rate or increase non-uniformity across the substrate.

[0006] Typically, the polishing pad is maintained at the desired surface roughness (and glazing is avoided) through a conditioning process using a pad conditioner. A polishing pad conditioner is used to remove unwanted buildup on the polishing pad and regenerate the surface of the polishing pad to the desired roughness. A typical pad conditioner includes an abrasive head, usually embedded with diamond abrasives, which can be scraped against the polishing pad surface to retexture the pad. Summary of the invention

[0007] In one aspect, an apparatus for chemical mechanical polishing includes: a platform having a surface to support a polishing pad; a carrier head that holds a substrate against the polishing surface of the polishing pad; a pad conditioner that presses an abrasive body against the polishing surface; an in-situ polishing pad monitoring system that includes an imager disposed above the platform to capture an image of the polishing pad; and a controller configured to receive the image from the monitoring system and generate a measurement of the surface roughness of the polishing pad based on the image.

[0008] Implementations may include one or more of the following features.

[0009] The controller may be configured to operate as a machine learning based image processing system and to input the image to the image processing system. The machine learning based image processing system may include a supervised learning module. The machine learning based image processing system may include a size reduction module to receive the image and output component values, and the controller may be configured to input the component values ​​of the image to the supervised learning module. The controller may be configured to input the image directly to the supervised learning module. The controller may be configured to operate the supervised learning module as an artificial neural network.

[0010] The controller may be configured to receive other data including values ​​of parameters, and may be configured to generate a measurement of the surface roughness of the polishing pad based on the image and the values ​​of the parameters. The parameters may be polishing control parameters, state parameters, measurements from sensors in the polishing system, or measurements of the polishing pad obtained by sensors external to the polishing system. The parameters may be a platen rotation rate, a slurry dispense rate, a slurry composition, a number of substrates after the polishing pad is changed, or a measurement of the surface roughness of the polishing pad obtained by an independent metrology station before the polishing pad is mounted on the platen.

[0011] The controller can be configured to at least one of: stop the conditioning process, or adjust a conditioning parameter based on the measurement of the polishing pad surface roughness.

[0012] The imager may be radially movable on the platform.The imager may be mounted on an arm that can swing laterally on the platform.

[0013] In another aspect, a polishing method includes: contacting a substrate with a polishing pad on a platform; generating relative motion between the substrate and the polishing pad; capturing an image of the polishing pad using an optical sensor; and generating a measurement of the surface roughness of the polishing pad by inputting the image into an image processing system based on machine learning.

[0014] Implementations may include one or more of the following features.

[0015] Training data including a plurality of pairs of training images and training values ​​may be received. The training data may be used to train a supervised learning algorithm in a learning-based image processing system. A conditioning process may be stopped or conditioning parameters may be adjusted based on the measurement of the surface roughness of the polishing pad.

[0016] Certain embodiments may include, but are not limited to, one or more of the following advantages. The roughness of the polishing pad may be determined using a non-contact technique, thus contamination and damage to the polishing pad may be avoided. The roughness of the polishing pad may be accurately and quickly determined, and the conditioning process may be adjusted appropriately. Wafer-to-wafer non-uniformity (WTWNU) may be reduced. The roughness may be determined using a non-contact technique, thus contamination of the polishing pad may be avoided.

[0017] The details of one or more implementations are set forth in the accompanying drawings and the description below.Other aspects, features, and advantages will be apparent from the description and drawings, and from the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a schematic partial cross-sectional side view of a chemical mechanical polishing system including an optical monitoring system configured to detect the surface roughness of a polishing pad.

[0019] Figure 2 is a schematic top view of a chemical mechanical polishing system.

[0020] Figure 3 A block diagram of an image processing system based on machine learning.

[0021] Like reference numbers in the various drawings indicate like elements. DETAILED DESCRIPTION

[0022] The chemical mechanical polishing process tends to reduce the surface roughness of the polishing pad, for example due to the glazing effect mentioned above. Conditioning can be used to restore the surface roughness. However, the degree of glazing of the pad and the degree to which the surface roughness is restored by the adjustment can be non-uniform across the polishing pad. Therefore, even after adjustment, the surface roughness of the polishing pad can be non-uniform. In addition, the conditioning technique can wear the pad at different rates across the pad, resulting in non-uniformity in pad thickness, and can leave periodic scratches or notches on the surface of the polishing pad. The surface roughness can be measured using contact techniques (such as profilometers), but this may introduce a risk of contamination, and it may be impractical to assemble the contact technique on the pad, given that equipment (conditioners, carriers, etc.) is already required. However, the polishing pad can be imaged, and the image can be fed to a trained machine learning model, which outputs a surface texture measurement result. The controller can then use this measurement result to adjust the conditioning process to achieve a target surface texture or improve the uniformity of the surface texture of the entire polishing pad.

[0023] As used herein, the term "surface texture" includes surface roughness, such as Ra, Rms, RSk, or Rp, and other irregularities in the polishing pad surface that are smaller than the normal groove or perforation pattern on the polishing pad, such as waviness. For example, assuming a groove depth of 20 mils, the surface texture may contain irregularities up to about 40 to 50 microns.

[0024] Figure 1 and Figure 2 An example of a polishing system 20 of a chemical mechanical polishing apparatus is illustrated. The polishing system 20 includes a rotatable disc-shaped platform 24 with a polishing pad 30 located on the platform 24. The platform 24 can be operated to rotate about an axis 25. For example, a motor 22 can rotate a drive shaft 28 to rotate the platform 24. The polishing pad 30 can be a dual-layer polishing pad having an outer layer 34 and a softer backing layer 32. The upper surface of the polishing pad 30 provides a polishing surface 36.

[0025] The polishing system 20 can include a supply port or combined supply rinse arm 39 to dispense a polishing liquid 38 , such as a slurry, onto the polishing pad 30 .

[0026] The polishing system 20 may also include a polishing pad conditioner 60 to abrade the polishing pad 30 to maintain the polishing surface 36 in a consistent abrasive state. The polishing pad conditioner 60 includes a base, an arm 62 that can sweep laterally over the polishing pad 30, and a conditioner head 64 connected to the base by the arm 64. The conditioner head 64 brings an abrasive surface (e.g., a lower surface of a disk 66 held by the conditioner head 64) into contact with the polishing pad 30 to condition the polishing pad 30. The abrasive surface may be rotatable, and the pressure of the abrasive surface against the polishing pad may be controllable.

[0027] In some embodiments, arm 62 is pivotally attached to the base and sweeps back and forth to move conditioner head 64 in an oscillating sweeping motion across polishing pad 30. The motion of conditioner head 64 may be synchronized with the motion of carrier head 70 to prevent collisions.

[0028] Vertical movement of the conditioner head 64 and pressure control of the conditioning surface on the polishing pad 30 may be provided by a vertical actuator 68 above or in the conditioner head 64, such as a pressurizable chamber positioned to apply downward pressure to the conditioner head 64. Alternatively, vertical movement and pressure control may be provided by a vertical actuator that raises and lowers the entire arm 62 and base of the conditioner head 64, or by a pivot connection that allows the arm 62 to control the angle of tilt, and thereby the height of the conditioner head 64 above the polishing pad 30.

[0029] The carrier head 70 is operable to hold the substrate 10 against the polishing pad 30. The carrier head 70 is suspended from a support structure 72 (e.g., a carousel or track) and is connected to a carrier head rotation motor 76 by a drive shaft 74 so that the carrier head can rotate about an axis 71. Optionally, the carrier head 70 can oscillate laterally, for example, on a slider on the carousel or track 72; or by rotational oscillation of the carousel itself. In operation, the platform rotates about its central axis 25, and the carrier head rotates about its central axis 71 and translates laterally across the top surface of the polishing pad 30. The carrier head 70 can include an elastic membrane 80 having a substrate mounting surface to contact the back side of the substrate 10, and a plurality of pressurizable chambers 82 to apply different pressures to different areas (e.g., different radial areas) on the substrate 10. The carrier head can also include a retaining ring 84 to hold the substrate.

[0030] The polishing system 20 includes an in-situ optical pad monitoring system 40 that generates a signal representative of the surface texture (e.g., surface roughness) of the polishing pad 30. The in-situ optical pad monitoring system 40 includes an imager 42, such as a camera positioned above the polishing pad, such as on a support arm 44. For example, the imager 42 can be a line scan camera, and the pad monitoring system 40 can be configured to generate a two-dimensional image from a plurality of measurements by the line scan camera as the polishing pad 30 is swept under the camera 40 due to the rotation of the platform 24. Alternatively, the imager 42 can be a two-dimensional camera. The imager 42 can have a field of view 43 of a portion of the surface 36 of the polishing pad 30. The camera can include a CCD array and optical components, such as a lens, to focus the imaging plane on the surface 36 of the polishing pad 30.

[0031] In some embodiments, the imager 42 is positioned in a fixed radial position and images a fixed radial area of ​​the polishing pad 30. In this case, the in-situ pad monitoring system 40 can generate measurements of surface texture (eg, surface roughness) at a fixed radial position on the polishing pad 30.

[0032] However, in some embodiments, the imager 42 can be moved laterally, for example, along a radius of the polishing pad 30. For example, referring to Figure 2 , the base 46 holding the support arm 42 can be configured to pivot so that the arm 42 swings across the polishing pad 30 (see arrow A) and carries the imager 42 in an arcuate path. As another example, the support arm 44 can be or include a linear track, and the imager 42 can be movable along the track by a linear actuator 46 (e.g., a stepper motor with a linear screw). By taking images of the polishing pad 30 at different radial regions, the in-situ pad monitoring system 40 can generate measurements of surface texture (e.g., surface roughness) at different radial locations on the polishing pad 30.

[0033] The controller 90 (e.g., a general purpose programmable digital computer) receives the images from the in-situ polishing pad monitoring system 40 and may be configured to generate from the images a measurement of the surface texture (e.g., surface roughness) of the polishing pad 30. In this context, the controller 90 (or the portion of the software that provides the surface texture measurement) may be considered to be part of the pad monitoring system 40. As described above, the surface roughness of the polishing pad 30 changes over time (e.g., during the course of polishing multiple substrates) due to the polishing and conditioning processes.

[0034] The controller 90 can also be configured to control the system of the pad conditioner 60 based on the value of the surface texture (e.g., surface roughness) received from the in-situ pad monitoring system 40. For example, when the measurement of the surface texture of the polishing pad 30 meets a threshold value, the controller 90 can stop the conditioning process. As another example, if the surface texture of the polishing pad meets another threshold value, the controller 90 can generate an alert to the operator of the polishing system 20 (e.g., the polishing or conditioning operation is not proceeding as expected).

[0035] As another example, if the in-situ pad monitoring system 40 generates measurements for surface texture (e.g., surface roughness) at different radial locations on the polishing pad 30 (relative to the rotation axis 25), the controller 90 may use the measurements to control the pad conditioner 60 to improve the uniformity of the surface texture (e.g., surface roughness). For example, the controller 90 may control the sweeping of the conditioner arm 62 to control the dwell time of the conditioner disk 64 in different radial regions on the polishing pad. For example, if it is desired to increase the surface roughness in a radial region, the dwell time may be increased, and if it is desired to decrease the surface roughness in a radial region, the dwell time may be decreased.

[0036] Reference Figure 3 , images from the in-situ pad monitoring system 40 are fed into a trained machine vision image processing system 100. The machine vision image processing system 100 is configured to output a value representing the texture (e.g., surface roughness) of a portion of the polishing surface 36 within the field of view 43 of the imager 42. The machine vision image processing system 100 may be implemented as part of the controller 90. The machine vision image processing system 100 may incorporate various machine learning techniques. For example, the machine vision image processing system 100 may include a neural network, but other approaches are also possible, such as a naive Bayes classifier or a support vector machine.

[0037] Figure 3The functional blocks that may be implemented for the machine learning based image processing system 100 are illustrated. These functional blocks may include an optional size reduction module 110 to perform size reduction of the image, and a supervised learning module 120 (shown as being implemented as a neural network). The supervised learning module 120 implements the functionality of a supervised learning algorithm to generate output surface texture (e.g., surface roughness) measurements based on the image (or size reduced data from the image). These functional blocks may be distributed across multiple computers.

[0038] The output of the supervised learning module 120 may be fed to a process control system 130, which may be implemented as part of the controller 90, to adjust the polishing process based on the surface texture measurements. For example, the process control system 130 may detect an adjustment endpoint and stop adjustment and / or adjust adjustment parameters (e.g., sweep profile, conditioner head pressure, etc.) during or between polishing processes based on the surface texture (e.g., surface roughness) measurements to reduce non-uniformity of the surface texture (e.g., surface roughness) of the polishing surface 36.

[0039] Assume that the machine learning module 120 is a neural network that includes a plurality of input nodes 122 for each principal component, a plurality of hidden nodes 124 (hereinafter also referred to as "intermediate nodes"), and an output node 126 that will generate a surface texture (e.g., surface roughness) measurement result. Typically, the hidden node 124 outputs a value of a nonlinear function of a weighted sum of the values ​​from the input nodes 122 connected to the hidden node.

[0040] For example, the output of hidden node 124 (designated as node k) may be represented as: where tanh is the hyperbolic tangent, a kx is the connection weight between the kth intermediate node and the xth input node (out of the M input nodes), and I M is the value at the Mth input node. However, other nonlinear functions may be used instead of tanh, such as the rectified linear unit (ReLU) function and its variants.

[0041] The optional size reduction module 110 reduces the image to a more limited number of component values ​​112, such as L component values. The neural network 120 includes an input node 122 for each component to which the image is reduced, for example, in the case where the size reduction module 110 generates L component values, the neural network 120 will include at least input nodes N1, N2 ... N L .

[0042] However, the supervised learning module 120 may optionally receive one or more inputs 114 in addition to the image or component values. The other input(s) 114 may include a measurement from another sensor in the polishing system, for example, a measurement of the temperature of the pad obtained by a temperature sensor, or a measurement of the slurry flow rate from a flow sensor. The other input(s) may include values ​​of polishing control parameters, for example, the platform rotation rate, the slurry flow rate, or the slurry composition. The polishing control parameter values ​​may be obtained from a polishing recipe stored by the controller 90. The other input(s) may include state parameters tracked by the controller, for example, an identification of the type of pad used (such as the manufacturer, brand name, composition of the pad, groove pattern, etc.), or the number of substrates polished after the pad was changed. The other input(s) may include measurements from sensors that are not part of the polishing system, for example, measurements of the surface texture (e.g., surface roughness) of the polishing pad obtained by an independent metrology station before the pad is mounted on the platform. This allows the supervised learning module 120 to take these other process or environmental variables into account when calculating the surface texture (e.g., surface roughness). Assuming that the supervised learning module 120 is a neural network, the neural network may include one or more other input nodes (eg, node 122a) to receive other data.

[0043] The architecture of the neural network 120 may vary in depth and width. For example, although the neural network 120 is shown as having a single column of hidden nodes 124, the neural network 120 may include multiple columns. The number of intermediate nodes 124 may be equal to or greater than the number of input nodes 122. The neural network may be a fully connected network or a convolutional network.

[0044] Before using the supervised learning module 120 for processing of, for example, a device wafer, the supervised learning module 120 needs to be configured.

[0045] As part of the configuration procedure, the supervised learning module 120 receives training data, which may include a plurality of training images and a plurality of training values, such as surface texture values, such as surface roughness values. Each reference image has a training value, ie, the training data includes pairs of images and training values.

[0046] For example, the images may be taken of various pad samples. Additionally, measurements of the surface roughness of the samples may be performed using metrology equipment such as a contact profilometer, an interferometer, or a confocal microscope. Each training image may thus be associated with a training value, which is the surface roughness of the sample from which the image was taken.

[0047] In some embodiments, the data store may store multiple sets of training data. Different sets of training data may correspond to different types of polishing pads, for example, different compositions and / or groove patterns. The supervised learning module 120 may receive a selected set of training data from an operator of a semiconductor manufacturing plant, for example, via a user interface.

[0048] The training of the supervised learning module 120 may be performed using conventional techniques. For example, for a neural network, training may be performed by back propagation using training images and training values. For example, when the neural network is operating in training mode, the reduced-size values ​​of the training images are fed to the corresponding input nodes N1, N2, ... N L , while the training value V is fed to the output node 126. This operation can be repeated for each pair of image and training value. In the case where the supervised learning module 120 receives inputs other than images or component values, the values ​​for these parameters can also be fed to the machine learning module as training data.

[0049] Once training has been performed, the trained instance of the supervised learning module can then be used, for example, as described above. That is, during processing of a substrate, images of the polishing pad and other parameter values ​​can be fed to the trained supervised learning module 120, which outputs a value of surface texture (e.g., surface roughness). The surface texture value (e.g., surface roughness) can then be used to control a conditioning operation, for example, as described above.

[0050] The in-situ polishing pad monitoring system can be used for various polishing systems. The polishing pad or the carrier head or both can be moved to provide relative motion between the polishing surface and the substrate. The polishing pad can be a round (or some other shape) pad fixed to the platform, a belt extending between a supply roller and a collection roller, or a continuous belt. The polishing pad can be fixed to the platform, gradually advanced on the platform between polishing operations, or continuously driven on the platform during polishing. The pad can be fixed to the platform during polishing, or there can be a fluid bearing between the platform and the polishing pad during polishing. The polishing pad can be a standard (e.g., polyurethane with or without filler) rough pad, soft pad, or fixed abrasive pad.

[0051] Additionally, while the foregoing focuses on monitoring during polishing, measurements of the polishing pad may be obtained before or after polishing the substrate (eg, while the substrate is being transferred to the polishing system).

[0052] The controller and its functional operations may be implemented in digital electronic circuitry, or in computer software, firmware or hardware, or a combination thereof. A controller that is "configured" to perform an operation has sufficient software, firmware or hardware to actually perform the operation, rather than merely being capable of being programmed or modified to perform the operation.

[0053] Embodiments may be implemented as one or more computer program products, i.e., one or more computer programs tangibly embodied in an information carrier (e.g., in a non-transitory machine-readable storage medium or in a propagating signal) to be executed by a data processing device (e.g., a programmable processor, a computer or multiple processors or computers) or to control the operation of a data processing device. Computer programs (also referred to as programs, software, software applications, or codes) may be written in any form of programming language (including compiled or interpreted languages) and may be deployed in any form, including as stand-alone programs or as modules, components, subroutines, or other units suitable for use in a computing environment. A computer program does not necessarily correspond to a file. A program may be stored in a portion of a file that stores other programs or data, in a single file dedicated to the program, or in multiple coordinated files (e.g., files storing portions of one or more modules, subroutines, or codes). A computer program may be deployed to be executed on one or more computers at one site or distributed across multiple sites and interconnected by a communication network.

[0054] The processes and logic flows described in this specification may be performed by one or more programmable processors executing one or more computer programs to operate on input data and generate output to perform functions. The processes and logic flows may also be performed by and the device may be implemented as a special purpose logic circuit system, such as an FPGA (field programmable gate array) or an ASIC (application specific integrated circuit).

[0055] Numerous embodiments of the present invention have been described. However, it should be understood that various modifications can be made without departing from the spirit and scope of the present invention. Accordingly, other embodiments are within the scope of the appended claims.

Claims

1. An apparatus for chemical mechanical polishing, comprising: a platform having a surface to support a polishing pad; a carrier head that holds a substrate against a polishing surface of the polishing pad; a pad conditioner that presses an abrasive body against the polishing surface; an in-situ polishing pad monitoring system comprising an imager disposed above the platform to capture an image of the polishing pad; as well as A controller is configured to receive the image from the monitoring system, receive other data including a value of a parameter, and generate a measurement of a polishing pad surface texture based on the image and the value of the parameter.

2. The apparatus of claim 1, wherein the parameter comprises a polishing control parameter, a status parameter, a measurement result from a sensor in the polishing system, or a measurement result of the polishing pad obtained by a sensor external to the polishing system.

3. The apparatus of claim 2, wherein the parameters include a platform rotation rate, a slurry dispensing rate, a slurry composition, a number of substrates after the polishing pad is changed, or a measurement of the surface roughness of the polishing pad obtained by an independent metrology station before the polishing pad is mounted on the platform.

4. The apparatus of claim 1, wherein the controller is configured to at least one of: stop a conditioning process, or adjust a conditioning parameter based on the measurement of the polishing pad surface texture.

5. The apparatus of claim 1, wherein the controller is configured to operate as a machine learning based image processing system and is configured to input the image to the image processing system.

6. The apparatus of claim 5, wherein the machine learning based image processing system comprises a supervised learning module.

7. The apparatus of claim 6, wherein the machine learning-based image processing system comprises a size reduction module to receive the image and output component values, and wherein the controller is configured to input the component values ​​of the image to the supervised learning module.

8. The apparatus of claim 6, wherein the controller is configured to input the image directly into the supervised learning module.

9. The apparatus of claim 6, wherein the controller is configured to operate the supervised learning module as an artificial neural network.

10. An apparatus for chemical mechanical polishing, comprising: a platform having a surface to support a polishing pad; a carrier head that holds a substrate against a polishing surface of the polishing pad; a pad conditioner that presses an abrasive body against the polishing surface; an in-situ polishing pad monitoring system, the in-situ polishing pad monitoring system comprising an imager disposed above the platform to capture an image of the polishing pad, wherein the imager is radially movable above the platform; as well as A controller is configured to receive the image from the monitoring system and generate a measurement of a polishing pad surface texture based on the image.

11. The apparatus of claim 10, wherein the imager is mounted on a pivotable arm to swing laterally above the platform.

12. A polishing method comprising: contacting the substrate with a polishing pad on a platform; generating relative motion between the substrate and the polishing pad; capturing an image of the polishing pad using an optical sensor; Receives additional data including the value of the parameter; as well as A measurement of the surface texture of the polishing pad is generated by inputting the image and the value of the parameter into a machine learning based image processing system.

13. The method of claim 12, wherein the parameter comprises a polishing control parameter, a state parameter, a measurement from a sensor in the polishing system, or a measurement of the polishing pad obtained by a sensor external to the polishing system.

14. The method of claim 12, wherein the parameters include a platform rotation rate, a slurry dispensing rate, a slurry composition, a number of substrates after the polishing pad is changed, or a measurement of the surface roughness of the polishing pad obtained by an independent metrology station before the polishing pad is mounted on the platform.

15. The method of claim 12, comprising receiving training data comprising a plurality of pairs of training images and training values, and using the training data to train a supervised learning algorithm in the learning-based image processing system. The method of claim 15 , wherein the training values ​​include surface texture values. The method of claim 16 , wherein the training values ​​include surface roughness values.

18. The method of claim 12, comprising stopping a conditioning process or adjusting a conditioning parameter based on the measurement of the polishing pad surface texture.

Citation Information

Patent Citations

  • Interconnected-multi-element-lattice polishing pad

    CN101367190A

  • Workpiece polishing method and workpiece polishing apparatus

    CN108340281A

  • Chemical mechanical polishing's polishing pad image detection system

    CN208514344U

  • Polishing pad adjusting device for cmp device

    JP1999277405A

  • In-situ chemical-mechanical planarization pad metrology using ultrasonic imaging

    US20040043521A1