Displacement measurement in semiconductor wafer processing

Abstract

When layers and features are added to an underlying substrate, a wafer starting as a flat surface during a semiconductor manufacturing process may be warped or curved. This warpage may be detected between manufacturing processes by rotating the wafer adjacent to the displacement sensor. Displacement sensors may generate displacement data relative to a baseline measurement to identify regions of the wafer that are curved upward or downward. The displacement data may then be mapped to a position on the wafer relative to the alignment feature. This mapping can then be used to adjust parameters in a subsequent semiconductor process, including adjusting how a carrier head on the polishing process holds the wafer or applies pressure to the wafer when polishing the wafer. The model may be trained to provide control signals for the polishing / cleaning process, or to generate metrology data.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application is a continuation-in-part of U.S. application No. 17 / 334,407 filed on May 28, 2021, which is incorporated herein by reference. Technical Field

[0002] The present disclosure generally relates to measuring the warpage of a wafer between semiconductor manufacturing processes. Specifically, the present disclosure describes measuring displacement data around a wafer to characterize the warpage / bow in the wafer, and then using the displacement data to control subsequent processes to compensate for the warpage / bow. Background Art

[0003] Integrated circuits are typically formed on a substrate by sequentially depositing conductive, semiconductive, or insulating layers on a silicon wafer. Various manufacturing processes use planarization of layers on a substrate between processing steps. For example, for some applications, such as polishing a metal layer to form vias, plugs, and / or lines in trenches of a patterned layer, the overlying layer is planarized until the top surface of the patterned layer is exposed. In other applications (e.g., planarizing a dielectric layer for photolithography), the upper layer is polished until a desired thickness is left above the lower layer.

[0004] Chemical mechanical polishing (CMP) is a common planarization method. This planarization method generally requires mounting the substrate on a carrier or polishing head. The exposed surface of the substrate is generally placed against a rotating polishing pad. The carrier head provides a controllable load on the substrate to push the substrate against the polishing pad. An abrasive polishing slurry is generally supplied to the surface of the polishing pad.

[0005] One problem in CMP and other semiconductor processes is maintaining a consistent surface on a wafer. Although the wafer may initially be a relatively flat silicon crystal, various semiconductor processes may deposit, grow, add, etch, remove, and / or otherwise place layers or features on the wafer. These additional layers and features may accumulate stress and cause stress to appear in the wafer. The stress may then cause the wafer to bend or warp. This warping may cause problems when polishing or cleaning the surface of the wafer, and the surface of the wafer may no longer be uniform. Summary of the invention

[0006] In some embodiments, a system includes a platform configured to support a wafer and rotate the wafer about a center of the wafer; and one or more displacement sensors positioned to measure displacement of the wafer as the wafer rotates.

[0007] In some embodiments, a method of measuring the displacement of a wafer between semiconductor manufacturing processes may include: rotating the wafer around the center of the wafer; measuring the displacement of the wafer at a predetermined distance from the center of the wafer as the wafer rotates to generate displacement data; and using the displacement data to control a polishing process or a cleaning process performed on the wafer.

[0008] In some embodiments, a non-transitory computer-readable medium may include instructions that, when executed by one or more processors, cause the one or more processors to perform operations including: receiving, for a wafer, displacement data generated using a displacement sensor, wherein the displacement data may indicate a bent or warped area on the wafer; identifying parameters of a carrier head corresponding to a location of the bent or warped area on the wafer, wherein the carrier head may be part of a chemical mechanical polishing (CMP) machine or a wafer cleaning machine; and adjusting the parameters of the carrier head to compensate for the bent or warped area on the wafer during a cleaning or polishing process.

[0009] In any embodiment, any and all of the following features may be implemented in any combination and are not limiting. The system may also include an alignment sensor positioned to detect a notch in the wafer as the wafer rotates. The one or more displacement sensors may include a laser displacement sensor; and the alignment sensor may include a through-beam sensor. The one or more displacement sensors may also be positioned to detect a notch in the wafer as the wafer rotates. The one or more displacement sensors may include a displacement sensor positioned above the wafer. The one or more displacement sensors may include a displacement sensor positioned below the wafer. The one or more displacement sensors may include a plurality of displacement sensors positioned along a radial line of the wafer. The one or more displacement sensors may include a plurality of displacement sensors positioned at different rotation angles around the wafer. The method / operation may also include measuring a displacement of a baseline wafer to generate baseline displacement data representing a flat wafer surface. Generating the displacement data may include determining a displacement of the wafer relative to the baseline displacement data. The method / operation may also include mapping the displacement data to a two-dimensional coordinate on the wafer based on the alignment data. The method / operation may also include mapping the displacement data to a rotation angle on the wafer based on the alignment data. The method / operation may also include identifying a displacement region on the wafer in the displacement data, and providing the displacement region to a polishing process or a cleaning process. Adjusting the parameters of the carrier head may include adjusting the angle at which the carrier head grips the wafer during a cleaning or polishing process. Adjusting the parameters of the carrier head may include adjusting the amount of pressure provided against a curved or warped area on the wafer during a cleaning or polishing process. Adjusting the amount of pressure may include identifying a chamber in the carrier head adjacent to the curved or warped area on the wafer; and adjusting the pressure in the chamber to compensate for the curved or warped area on the wafer. Adjusting the amount of pressure may include adjusting the pressure used by a brush to clean the curved or warped area on the wafer during a cleaning process. The displacement sensor may be used to generate displacement data by rotating the displacement sensor around the wafer.

[0010] In some embodiments, a system may include: one or more displacement sensors positioned to measure displacement of a wafer as the wafer rotates on a platform; a cleaning or polishing chamber configured to clean or polish the wafer; and a computer system configured to perform operations that may include: receiving displacement data generated using the one or more displacement sensors for the wafer, wherein the displacement data may indicate displacement of the wafer relative to the one or more displacement sensors; providing the displacement data to a model trained using a displacement data set; in response to receiving the displacement data as input, generating a control signal for the cleaning or polishing chamber based on an output from the model; and using the control signal to control a cleaning or polishing process for the wafer.

[0011] In some embodiments, a method may include: receiving, for a wafer, displacement data generated using a displacement sensor, wherein the displacement data may indicate displacement of the wafer relative to the displacement sensor; providing the displacement data to a model trained using a metrology data set and a corresponding displacement data set; and in response to receiving the displacement data as input to the model, generating metrology data for the wafer based on output from the model.

[0012] In some embodiments, a non-transitory computer-readable medium may include instructions that, when executed by one or more processors, cause the one or more processors to perform operations including: receiving, for a wafer, displacement data generated using one or more displacement sensors, wherein the displacement data may indicate a displacement of the wafer relative to the one or more displacement sensors; providing the displacement data to a model trained using a displacement data set; and in response to receiving the displacement data as input, generating a control signal for cleaning or polishing a chamber based on an output from the model.

[0013] In any embodiment, any and all of the following features may be implemented in any combination and are not intended to be limiting. Controlling a cleaning or polishing process for a wafer may include selecting a cleaning or polishing process from a plurality of different cleaning or polishing processes, wherein the cleaning or polishing process may have a shorter run time than at least one of the plurality of different cleaning or polishing processes. Controlling a cleaning or polishing process for a wafer may include controlling the rotation speed of a platform supporting the wafer during the cleaning or polishing process. Controlling a cleaning or polishing process for a wafer may include controlling the rotation speed of a cleaning or polishing head during the cleaning or polishing process. Controlling a cleaning or polishing process for a wafer may include controlling the flow of a slurry used in the cleaning or polishing process. Controlling a cleaning or polishing process for a wafer may include controlling the pressure applied to a partition in a cleaning or polishing head during the cleaning or polishing process. Displacement data may indicate a curved or warped zone on a wafer. The method / operation may also include using metrology data to control a cleaning or polishing process for a wafer. The metrology data may include the thickness of a film or feature on a wafer. The method / operation may also include rotating the wafer about the center of the wafer, measuring the displacement of the wafer at a predetermined distance from the center of the wafer as the wafer rotates to generate displacement data, and / or measuring the displacement of a baseline wafer to generate baseline displacement data representing a flat wafer surface. Generating the displacement data may include determining the displacement of the wafer relative to the baseline displacement data. The method / operation may also include mapping the displacement data to two-dimensional coordinates on the wafer based on the alignment data, or mapping the displacement data to a rotation angle on the wafer based on the alignment data. The method / operation may also include identifying a displacement region on the wafer in the displacement data, and providing the displacement region to a polishing or cleaning process. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] A further understanding of the nature and advantages of the various embodiments may be realized by reference to the remainder of the specification and the drawings, wherein the same reference numerals are used in the several drawings to represent similar components. In some cases, a sub-label is associated with a reference numeral to indicate one of multiple similar components. When reference is made to a reference numeral without specifying the existing sub-label, it is intended to refer to all such multiple similar components.

[0015] Figure 1 A schematic cross-sectional view of an example of a polishing apparatus according to some embodiments is shown.

[0016] Figure 2 Shown are optical spectra measured from an in-situ optical monitoring system according to some embodiments.

[0017] Figure 3 A path of a sequence of spectral measurements on a substrate according to some embodiments is shown.

[0018] Figure 4 A system for measuring wafer displacement according to some embodiments is shown.

[0019] Figure 5A A displacement sensor positioned at the edge of a wafer is shown in accordance with some embodiments.

[0020] Figure 5B A displacement sensor is shown positioned to measure displacement of a bottom edge of a wafer in accordance with some embodiments.

[0021] Figure 5C A plurality of displacement sensors are shown in accordance with some embodiments.

[0022] Figure 5D Multiple displacement sensors positioned at different rotational angles are shown in accordance with some embodiments.

[0023] Figure 6 A three-dimensional graph of displacement data is shown in accordance with some embodiments.

[0024] Fig. 7A A baseline wafer that may be measured by a displacement sensor is shown according to some embodiments.

[0025] Figure 7B A convex wafer is shown according to some embodiments.

[0026] Figure 7C A concave wafer is shown according to some embodiments.

[0027] Figure 8 A graph of displacement data for a warped wafer is shown in accordance with some embodiments.

[0028] Fig. 9 Flowcharts illustrating different processes that may be performed as part of a semiconductor manufacturing process in accordance with some embodiments.

[0029] Fig.10 A flow chart of a method for measuring displacement of a wafer during a semiconductor manufacturing process according to some embodiments is shown.

[0030] Fig.11 A block diagram of a process for training a model according to some embodiments is shown.

[0031] Fig.12 A block diagram of a system for controlling a cleaning or polishing process using a trained model is shown in accordance with some embodiments.

[0032] Fig.13 A block diagram of a process for using a model to generate control signals for a cleaner polishing process is shown in accordance with some embodiments.

[0033] Fig.14 A block diagram illustrating how a model may be used to generate metering data according to some embodiments.

[0034] Fig.15 A flow chart of a method for training and using a model with displacement data according to some embodiments is shown.

[0035] Fig.16 An exemplary computer system is shown in which various embodiments may be implemented. DETAILED DESCRIPTION

[0036] Figure 1 An example of a polishing apparatus 100 is shown. The polishing apparatus 100 may include a rotatable dish-shaped platform 120 on which a polishing pad 110 may be mounted. The platform may be operable to rotate about an axis 125. For example, a motor 121 may rotate a drive shaft 124 to rotate the platform 120. The polishing pad 110 may be a two-layer polishing pad having an outer polishing layer 112 and a softer backing layer 114.

[0037] The polishing apparatus 100 may include a port 130 for dispensing a polishing liquid 132, such as a slurry, onto the polishing pad 110. The polishing apparatus may also include a polishing pad conditioner for grinding the polishing pad 110 to maintain the polishing pad 110 in a consistent grinding state.

[0038] The polishing apparatus 100 may include at least one carrier head 140. The carrier head 140 may be operable to hold the substrate 10 against the polishing pad 110. The carrier head 140 may have independent control over polishing parameters (eg, pressure) associated with each respective substrate.

[0039] In particular, the carrier head 140 may include a retaining ring 142 to hold the substrate 10 beneath the flexible membrane 144. The carrier head 140 may also include a plurality of independently controllable pressurizable chambers defined by the membrane, for example, three chambers 146a-146c, which may apply independently controllable pressures to associated zones on the flexible membrane 144 and, therefore, on the substrate 10. Although for ease of illustration, the carrier head 140 may include a plurality of independently controllable pressurizable chambers, for example, three chambers 146a-146c, defined by the membrane, which may apply independently controllable pressures to associated zones on the flexible membrane 144 and, therefore, on the substrate 10. Figure 1 Only three chambers are shown in FIG, but there may be one or two chambers, or four or more chambers, for example, five chambers in other embodiments.

[0040] The carrier head 140 can be suspended from a support structure 150 (e.g., a turntable or track) and can be connected to a carrier head rotation motor 154 via a drive shaft 152 so that the carrier head can rotate about an axis 155. Optionally, the carrier head 140 can oscillate laterally, for example, on a slide on the turntable 150 or track; or by rotation of the turntable itself. In operation, the platform can rotate about its central axis 125, and the carrier head can rotate about its central axis 155 and / or translate laterally across the top surface of the polishing pad.

[0041] Although only one carrier head 140 is shown, more carrier heads may be provided to hold additional substrates so that the surface area of ​​the polishing pad 110 may be efficiently used. The polishing apparatus may also include an in-situ monitoring system 160. The in-situ monitoring system may generate a time-varying sequence of values ​​that depends on the thickness of the layer on the substrate. The in-situ monitoring system 160 may include an optical monitoring system. In particular, the in-situ monitoring system 160 may measure a sequence of spectra of light reflected from the substrate during polishing.

[0042] Optical access 108 through the polishing pad can be provided by including an aperture (i.e., a hole extending through the pad) or a solid window 118. The solid window 118 can be secured to the polishing pad 110, for example, as a plug that fills a hole in the polishing pad. The solid window can alternatively be molded into or adhesively secured to the polishing pad, although in some embodiments, the solid window can be supported on the platform 120 and protrude into a hole in the polishing pad.

[0043] The optical monitoring system 160 may include a light source 162, a light detector 164, and circuitry 166 for sending and receiving signals between a remote controller 190 (e.g., a computer) and the light source 162 and the light detector 164. One or more optical fibers may be used to transmit light from the light source 162 to an optical inlet in the polishing pad and to transmit light reflected from the substrate 10 to the detector 164. For example, a bifurcated optical fiber 170 may be used to transmit light from the light source 162 to the substrate 10 and back to the detector 164. The bifurcated optical fiber may include a trunk 172 positioned proximate to the optical inlet, and two branches 174 and 176 connected to the light source 162 and the detector 164, respectively.

[0044] In some embodiments, the top surface of the platform can include a recess 128 into which an optical head 168 is mounted, which holds one end of the trunk 172 of the bifurcated fiber. The optical head 168 can include a mechanism for adjusting the vertical distance between the top of the trunk 172 and the solid window 118.

[0045] The output of the circuit system 166 may be a digital electronic signal that is transmitted through a rotary coupling 129 (e.g., a slip ring) in the drive shaft 124 to a controller 190 for the optical monitoring system. Similarly, the light source may be turned on or off in response to a control command in a digital electronic signal transmitted from the controller 190 through the rotary coupling 129 to the optical monitoring system 160. Alternatively, the circuit system 166 may communicate with the controller 190 via wireless signals.

[0046] The light source 162 may be operable to emit ultraviolet (UV), visible light, or near infrared (NIR) light. The light detector 164 may be a spectrometer. A spectrometer is an optical instrument for measuring the intensity of light over a portion of the electromagnetic spectrum. For example, a grating spectrometer may be used. Common outputs of a spectrometer may include light intensity that varies with wavelength (or frequency). Figure 2 An example of a measured spectrum 200 of intensity as a function of wavelength is shown.

[0047] As described above, the light source 162 and the light detector 164 may be connected to a computing device, such as a controller 190, which is operable to control its operation and receive its signals. The computing device may include a microprocessor located near the polishing device. For example, the computing device may be a programmable computer. Regarding control, the computing device may, for example, synchronize the activation of the light source and the rotation of the platform 120. A display 192 (e.g., an LED screen) and a user input device 194 (e.g., a keyboard and / or a mouse) may be connected to the controller 190.

[0048] In operation, the controller 190 may receive, for example, a signal carrying information describing the spectrum of light received by the light detector for a particular flash of the light source or time frame of the detector. Thus, this spectrum may be a spectrum measured in situ during polishing. Without being limited to any particular theory, the spectrum of light reflected from the substrate 10 may evolve as polishing proceeds due to changes in the thickness of the outermost layer, thus producing a sequence of time-varying spectra.

[0049] The optical monitoring system 160 can be configured to generate a sequence of measured spectra, or "spectral data," at a measurement frequency. Relative motion between the substrate 10 and the optical inlet 108 results in the measurement of spectra in the sequence at different locations on the substrate 10. In some embodiments, the light beam generated by the light source 162 can be generated from a light source that rotates with the platform 120 (by Figure 3 The point indicated by the arrow R in the figure appears. Figure 3 As shown, in such an embodiment, relative motion between the substrate 10 and the optical inlet 108 can result in a spectrum being measured at a position 300 in a path across the substrate 10. In some embodiments, only one spectrum is measured each time the platform is rotated. In addition, in some embodiments, the emission point of the light beam is fixed, and measurements are taken only when the optical inlet 108 is aligned with the light beam.

[0050] The CMP process represents one of many different processes that can be performed during the manufacture of semiconductor wafers. As different layers and features are added to the substrate, the CMP process can be used to planarize the wafer and remove specific amounts of material between stages. Ideally, the wafer would start as a completely flat silicon crystalline substrate. However, as layers and features are added to the substrate, this additional material adds stress to the structure and can cause warping or bending in the wafer. This warping can cause problems during the CMP process, as the wafer can be polished unevenly, resulting in layers that do not have uniform thickness after the CMP process is completed.

[0051] Embodiments described herein solve these and other technical problems by characterizing the vertical displacement of a wafer between semiconductor manufacturing processes. The displacement measurement can be performed using a displacement sensor around the circumference of the wafer. The resulting displacement data can then be used to control the polishing process, cleaning process, and / or the angle or orientation of the wafer for handling. This allows subsequent semiconductor processes to compensate for the warpage of the wafer.

[0052] Figure 4 A system 400 for measuring wafer displacement according to some embodiments is shown. The system 400 may include a platform 412 configured to support a wafer 404 and rotate the wafer 404 about a central axis 414 of the wafer 404. For example, an operator or a robot may place the wafer 404 on the platform 412 prior to placement in a CMP machine. The platform 412 may then rotate the wafer 404 about the central axis 414. Various sensors may measure properties of the wafer 404 during rotation.

[0053] In some embodiments, the system 400 may include an alignment sensor. The alignment sensor may be implemented using a through-beam sensor configured to pass a beam of light between the first component 401 and the second component 403. The alignment sensor may be configured to detect when the beam is interrupted between the first portion 401 and the second portion 403. The alignment sensor may therefore be used in conjunction with the rotation of the wafer 404 provided by the platform 412 to align the wafer 404 to a predefined orientation. For example, the wafer 404 may include a notch 416 or other edge feature along the outer periphery of the wafer 404. Typically, when the wafer 404 rotates around the central axis 414, the wafer 404 will block the beam between the first component 401 and the second component 403. However, when the position of the notch 416 along the edge of the wafer 404 passes over the first component 401 of the alignment sensor, the beam may be detected by the second component 403, thus indicating the position of the notch 416.

[0054] Once the location of the notch 416 is identified in the wafer 404, the system 400 can optionally rotate the wafer 404 to a predetermined orientation. For example, in some cases, the system 400 can stop the rotation of the wafer 404 when the notch 416 is located. In other cases, the system 400 can continue to rotate the wafer 404 until a predetermined rotation angle is reached. This process performed by the system 400 can be used to ensure that the wafer 404 is rotated to a known orientation before moving to a subsequent machine in the semiconductor manufacturing process. For example, some systems can use a robotic arm to carry the wafer 404 as the wafer 404 moves between machines. Before being placed on the polishing pad of a CMP machine (or any other semiconductor manufacturing machine), the wafer 404 can be rotated to a known position. As described below, this alignment and rotational orientation of the wafer 404 can also be used to identify the location of the bend or warp on the wafer 404.

[0055] The system 400 may be implemented at least in part by a machine called an aligner. Specifically, the aligner may be configured to rotate the wafer 404 until the notch 416 is located. In some embodiments, the aligner may be modified to also perform distance or displacement measurements on the wafer 404 while the wafer 404 is rotating. The system 400 may further include a displacement sensor 406. As used herein, the term "displacement sensor" may include any sensor configured to measure the distance of a surface from the sensor. As an example, some embodiments may use a laser displacement sensor. The laser displacement sensor may emit one or more laser beams directed at the surface of the wafer 404 and receive reflections from the wafer 404. The displacement sensor 406 may then calculate the distance between the displacement sensor 406 and the surface of the wafer 404 on which the laser beam is directed. When the wafer 404 rotates around the central axis 414, the displacement sensor 406 may continue to sample distance measurements relative to the surface of the wafer 404.

[0056] When modifying the aligner, the system 400 may also include a computer system that controls and / or records displacement data received from the displacement sensor 406. The computer system may include one or more processors (e.g., microprocessors, microcontrollers, digital controllers, FPGAs, programmable hardware, etc.) configured to control the operation of the displacement sensor 406 and record the displacement data. The one or more processors may be an integral part of the system 400. Alternatively, the one or more processors may be part of a separate computer system that communicates with the displacement sensor 406 via a wired or wireless interface. Instructions for controlling the displacement sensor 406 may be stored on one or more memory devices, such as non-transitory computer-readable media, and these instructions may cause the one or more processors to perform operations, such as receiving and processing displacement data, rotating the wafer 404, repackaging the displacement data, and the like. Fig.11An example of a computer system is described in .

[0057] The displacement sensor may be positioned to measure the displacement of the wafer 404 as the wafer 404 rotates. Measuring the displacement of the wafer 404 may generate displacement data. The displacement data may include a distance measurement from the displacement sensor 406 to the wafer 404. Alternatively, the displacement data may include the displacement of the wafer 404 from a baseline position. For example, the baseline position of the wafer 404 may be determined by measuring displacement data of a bare silicon substrate or other wafer known to be flat as a reference. If the wafer 404 is bent downward such that the displacement measurement is long compared to the baseline displacement, the displacement data may include a negative value. If the wafer 404 is bent upward such that the displacement measurement is short compared to the baseline displacement, the displacement data may include a corresponding positive value.

[0058] Figure 4 Examples of systems 400 include a system 400 that combines an aligner and a displacement sensor 406. However, other systems may omit the aligner and provide a dedicated displacement sensor with a platform that rotates the wafer 400, rather than modifying the aligner to include the displacement sensor 406. In other embodiments, the displacement sensor 406 may itself rotate about the central axis 414 while the wafer 404 remains stationary. In these embodiments, the displacement sensor itself may take the place of the aligner and perform alignment functions using the displacement sensor 406. For example, the displacement sensor 406 may collect displacement data and detect the location of the notch 416.

[0059] Figure 5A A displacement sensor 506a is shown positioned at the edge of a wafer 504 in accordance with some embodiments. Figure 4 In an example of , the displacement sensor 506 may be aimed at the outer edge of the wafer. For example, the outer edge of the wafer may include the outermost 1 cm of the wafer 504, so that the displacement measurement characterizes the displacement along the outer edge of the wafer 504. This position of the displacement sensor 506a may provide specific technical advantages. Specifically, the bending or warping of the wafer 504 may be most obvious at the edge. Since the wafer 504 is mounted at the central portion of the wafer 504 on the platform, the maximum displacement may be measured along the edge of the wafer 504. In addition, when aiming at the edge of the wafer, the displacement sensor 506a may also perform the function of an aligner. Specifically, the displacement sensor 506a may continue to sample the displacement measurement from the wafer 504. When a notch is encountered in the wafer 504, the displacement measurement may exceed the negative threshold distance, and the displacement sensor 506a may interpret this measurement as the location of the notch. Therefore, the displacement sensor 506a can determine the orientation and displacement data of the wafer 504.

[0060] Figure 5B5 shows a displacement sensor 506b positioned to measure the displacement of the bottom edge of the wafer 504 according to some embodiments. The displacement sensor 506b may be placed below the wafer 504 instead of Figure 5A 504. This location of the displacement sensor 506b can provide specific technical advantages. Specifically, when a robotic arm is used to move the wafer 504 to / from the system, the displacement sensor 506b positioned below the wafer 504 will generally not interfere with the robotic arm moving the wafer 504 when the robotic arm lifts the wafer 504. Another technical advantage provided by this arrangement relates to the effect of the light-emitting displacement sensor on the semiconductor device. Typically, the devices on the wafer can be placed on the top side of the wafer 504 and exposed. By placing the displacement sensor 506b below the wafer 504, the system can avoid shining lasers or other high-intensity light sources onto these semiconductor devices. This can be advantageous when these semiconductor devices are particularly sensitive to high-intensity light.

[0061] Figure 5C Multiple displacement sensors 506c, 508c are shown in accordance with some embodiments. In this example, multiple displacement sensors 506c, 508c may be positioned on a radial line 510 extending beyond the center of the wafer 504. This allows the system to measure displacement data at multiple points at each rotation angle. Figure 5C The example shows two displacement sensors 506c, 508c, but this number of displacement sensors is provided by way of example only and is not intended to be limiting. Other embodiments may include more than two displacement sensors. With multiple displacement sensors measuring displacement data along radial line 510, the system can generate displacement maps for multiple locations on the wafer 504 to produce a displacement map for the entire surface of the wafer 504 rather than just the edge of the wafer. Prior to the present disclosure, a metrology station was required that performed a laser scan of the entire surface of the wafer 504 in order to generate an accurate view of many different aspects of the wafer 504. However, operation of the metrology station is expensive and time consuming. As shown in Figure 5C As depicted in FIG. 5 , providing a plurality of displacement sensors 506 c , 508 may utilize a low cost, high speed process to replace at least a portion of the operations performed by a metrology station for providing surface mapping of displacement data for the wafer 504 .

[0062] Figure 5DA plurality of displacement sensors 506d, 508d positioned at different rotation angles according to some embodiments are shown. Rather than placing displacement sensors 506d, 508d along the same radial line at different radial distances, some embodiments may position displacement sensors 506d, 508d at the same radial distance separated by a known rotation angle 512. In the event of a failure of one of the displacement sensors 506d, 508d, these embodiments may provide repeated displacement data of the position as a backup. These embodiments may also use repeated displacement data to characterize, calibrate, and / or detect failures in the displacement sensors 506d, 508d. Note that more than two displacement sensors may also be used without limitation. Another technical advantage provided by aligning a plurality of displacement sensors 506d, 508d at the same radial distance is that parallel data of the wafer 504 is collected. For example, by placing a plurality of displacement sensors 506d, 508d about 180° apart on the wafer 504, the entire wafer perimeter may be measured in a half rotation rather than a full rotation. Similarly, by placing four displacement sensors approximately 90° apart on wafer 504, the entire wafer perimeter can be measured in a quarter rotation rather than a full rotation. These configurations can therefore reduce the overall time required to collect displacement data by 50%, 25%, etc.

[0063] exist FIG. 5A to FIG. 5B The combinations of different numbers and / or positions of displacement sensors shown in the figures may be used in any combination and are not limited. For example, some embodiments may include multiple displacement sensors at different rotation angles and / or different radial distances on the top side of the wafer, while optionally including additional displacement sensors at different rotation angles and / or different radial distances on the bottom side of the wafer. When using multiple displacement sensors, some embodiments may also use different sensor types or settings. For example, some embodiments may use multiple laser displacement sensors, each operating at a different optical frequency. Different frequencies may reflect from the wafer in different ways, so using different optical frequencies for different sensors can improve the accuracy of the overall displacement data measurement. In addition, when using multiple displacement sensors, some embodiments may use different sensor types. For example, a laser displacement sensor may be used in one location, while an ultrasonic displacement sensor may be used in another location.

[0064] Figure 6 A three-dimensional (3D) graph 600 of displacement data according to some embodiments is shown. Graph 600 includes distance measurements relative to a default or baseline displacement of a wafer. Positions on the wafer may be mapped into Cartesian coordinates on a millimeter scale. These coordinates may be assigned based on alignment data for the notch. For example, in the (x, y) plane of graph 600, a notch position may be assigned a coordinate position of (0, 150). Thus, once the alignment position in the displacement data has been captured, some embodiments may rotate the displacement data to conform to the coordinate position based on the notch position.

[0065] In this example, the displacement data may be stored as a vertical distance relative to a baseline position. For example, when no bowing or warping occurs, the ring 602 may represent the baseline position of the wafer. The data points 604 representing distances measured above the baseline position may be represented as positive displacements. These data points 604 may correspond to locations on the wafer that have upward bowing or warping relative to the center of the wafer. The data points 606 representing distances measured below the baseline position may be represented as negative displacements. These data points 606 may correspond to locations on the wafer that have downward bowing or warping relative to the center of the wafer.

[0066] As the wafers rotate on the platform, displacement data for each wafer may be stored. In addition, displacement data for each displacement sensor in the system that measures the wafer may be stored. The displacement data from the various displacement sensors may be combined for a single wafer to generate a 3D map of displacements across the surface of the wafer. As described below, this displacement data may then be communicated to other machines in the semiconductor processing pipeline and used to control operations on those machines.

[0067] Fig. 7A A baseline wafer 704 is shown that can be measured by a displacement sensor 706 according to some embodiments. The baseline wafer 704 can be an ideal flat wafer in the z-plane. In other words, the baseline wafer 704 can generate displacement data that shows a constant baseline displacement 720 aligned with the surface of the wafer 704. The displacement data received from the baseline wafer 704 can be used to characterize the relative displacement of other subsequent wafers measured by the system. The baseline wafer 704 can be measured as part of a calibration or initialization process for the system. Subsequent measured displacement data from subsequent wafers can be extracted from the displacement data from the baseline wafer. Although in Fig. 7A Although not explicitly shown in FIG. 1 , embodiments including multiple displacement sensors may store baseline displacement data for each sensor. Displacement data measured from subsequent wafers may be extracted from the baseline displacement data for each corresponding displacement sensor.

[0068] Figure 7B A convex wafer 704b is shown according to some embodiments. The convex profile indicates that the wafer 704b is bent or warped downward at the edge of the wafer 704b relative to the center of the wafer. The displacement 722 may be greater than the baseline displacement 720, resulting in a negative value when the displacement 722 is subtracted from the baseline displacement 720. Similarly, Figure 7C A concave wafer 704c is shown in accordance with some embodiments. The concave profile indicates that the wafer 704c is bowed or warped upward at the edge of the wafer 704c relative to the center of the wafer. The displacement 724 may be less than the baseline displacement 720, resulting in a positive value when the displacement 724 is subtracted from the baseline displacement 720.

[0069] Figure 8 A graph 800 of displacement data for a warped wafer is shown according to some embodiments. Rather than being mapped to 2D coordinates on a 3D axis, the displacement data may be mapped to a 2D axis, where the x-axis indicates the rotation angle and the y-axis indicates the displacement relative to a baseline. Mapping to the rotation angle may also be based on alignment data to determine the origin of the rotation. Displacement data 802 corresponds to a wafer that may include concave and / or convex portions. This type of wafer may be characterized as a "potato chip" shape with up and down ripples along the edge of the wafer. Displacement data 804 corresponds to a wafer in which one side is curled or warped upward. Graph 800 shows how displacement data for any type of warpage or bending condition in a wafer may be characterized by the system described herein.

[0070] Map to Figure 6 2D coordinates, or mapping to Figure 8 The angle rotation in can be used to control subsequent processes. For example, Figure 6 The 2D coordinates of the wafer can be used to identify 2D areas in the wafer that should be polished more heavily, cleaned more heavily, or require different pressures during the polishing process. For example, these different pressures can include lower pressures in some areas and higher pressures in other areas to compensate for irregularities in the wafer represented in the displacement data. Continuous areas of displacement on the wafer can be grouped together into regions and mapped to specific pressurized chambers in the CMP carrier head. These areas on the wafer can be identified in the mapping data described above.

[0071] Fig. 9 A flowchart 900 of different processes that may be performed as part of a semiconductor manufacturing process according to some embodiments is shown. Note that prior to beginning flowchart 900, the wafer may have been processed using various semiconductor machines, such as machines that perform various deposition processes, etching processes, and the like. After these processes have been performed, a robotic arm may move the wafer onto a wafer aligner / displacement measurement system 902. This system 902 may be implemented using system 400 described above. The wafer aligner / displacement measurement system 902 may generate displacement data 908.

[0072] The robot arm may move the wafer to the CMP machine 904 to perform the CMP process. The displacement data 908 may be sent to the handling control 910 and / or the endpoint control 912 of the CMP machine 904. The displacement data may then be used to control the polishing process performed on the wafer. Specifically, the CMP machine 904 may include a carrier head that grabs or holds the wafer. The carrier head may be associated with a plurality of parameters. These parameters may control the angle at which the wafer is grabbed by the carrier head, and / or the amount of pressure applied to the wafer at various locations. These parameters may be adjusted on the carrier head to compensate for curved or warped areas on the wafer during the cleaning or polishing process described below.

[0073] For example, the displacement data 908 may be used to control how the handling control 910 grips the wafer. For example, a flat wafer may be gripped by the carrier head with more force than a curved wafer. The CMP machine may receive the displacement data and adjust the gripping force applied to the wafer based on the amount of displacement identified in the displacement data. Additionally, if the wafer is curved along a portion of the circumference of the wafer, the handling control 910 may grip the wafer at a slight angle so that the curved portion is away from the wafer. Figure 1 904. In the embodiment of the present invention, the carrier head 140 on the CMP machine 100 shown in FIG. 904 is slightly more protruding than the rest of the wafer. In addition, the chambers 146 can be pressurized based on the displacement data 908. The specific chambers 146 behind the warped portion of the wafer can be pressurized more than the other chambers so that they push the warped portion to align with the rest of the wafer or apply more pressure during the polishing process. This can generate a flat surface on the wafer when the wafer is pushed against the polishing pad in the CMP machine 904. Because the orientation of the wafer is known from the alignment procedure, the displacement data can be mapped to the specific chambers 146 in the carrier head 140 above the warped portion of the wafer. These chambers can be pressurized more or less based on the positive / negative displacement data associated with each corresponding area of ​​the wafer.

[0074] The displacement data 908 may also be used by endpoint control 912 of the CMP process performed by the CMP machine 904. Figures 1 to 3When polishing the process described in , the real-time thickness of the top film on the wafer can be measured. For example, a flat wafer can be polished faster and with more force than a curved or warped wafer. Without access to displacement data, the polishing process typically uses an average or worst-case polishing speed / pressure to accommodate flat and warped wafers. However, when a wafer can be characterized as flat or warped using displacement data, the speed / pressure applied during the polishing process can be adjusted accordingly. The CMP machine can use the displacement data to determine whether the wafer is flat, and if the wafer is flat, the CMP machine can then increase the speed / pressure of the polishing process. If the wafer is not flat, the CMP machine can reduce the speed or pressure of the polishing process. This can result in a 5%-10% increase in the throughput of the wafer throughout the polishing process. Displacement data can also be provided to the CMP machine 904 to allow flat and warped wafers to be polished to have similar desired post-polishing characteristics. For example, a curved portion of a wafer can be polished more to generate an overall flat surface on the wafer compared to a flat portion of a wafer. Some embodiments can also adjust other parameters of the polishing process based on displacement data. For example, a substantially flat wafer may be polished faster using a higher temperature or a more abrasive slurry, while a warped wafer may be polished more finely using a less abrasive slurry, or vice versa. In addition, while a wafer is being polished, temperatures and chemical additives may be applied to different zones on the wafer. For example, a warped portion of a wafer may use a first chemical additive to the slurry and / or temperature, while a flat portion of the wafer may use a second chemical additive and / or temperature.

[0075] The measurement of the reflected wavelength can be affected by the displacement of the wafer from the polishing pad. The displacement data 908 can be used to compensate for locations on the wafer that may bend or warp, and therefore may result in inaccurate thickness measurements of the polished surface. For example, positive displacement data indicating that the wafer is bent toward the polishing pad can be added or subtracted from the thickness measurement by a proportional amount based on the effect of the displacement on the reflected wavelength received by the spectrometer. Ultimately, the endpoint can be adjusted to polish the warped portion of the wafer more than the rest of the wafer to produce an overall flat surface on the wafer for subsequent steps in the semiconductor manufacturing pipeline.

[0076] Next, the wafer may be moved to a cleaner 906 or a wafer cleaning machine configured to clean the debris remaining from the wafer after the polishing process. The displacement data 908 may be transferred to the handling control 914 of the cleaner 906 and / or the software implementing the cleaning recipe 916. The cleaner 906 may include a carrier head similar to the carrier head described above for the CMP machine 904. Next, the displacement data 908 may also be used to control the cleaning process performed on the wafer. For example, the handling control 914 may be changed to differently grasp the wafer based on the displacement data so that the wafer is more uniformly oriented for the cleaning process, as described above for the polishing process. In addition, the cleaning recipe 916 may be changed based on the warpage / bending profile of the wafer. Based on the relative positive / negative values ​​in the displacement data, the portion of the wafer corresponding to the warpage area may be cleaned more or less. For example, the pressure applied by the brush to the wafer may be adjusted using the positive / negative displacement data to apply more or less pressure to the protrusion area of ​​the wafer in the cleaning machine.

[0077] Fig.10 A flow chart 1000 of a method for measuring displacement of a wafer during a semiconductor manufacturing process according to some embodiments is shown. The method may include rotating a wafer around a center of the wafer (1002). This rotation may be performed while the wafer is supported on a platform, as described above in Figure 4 This rotation can also be used simultaneously to locate and align the notches on the wafer using alignment sensors.

[0078] The method may further include measuring the displacement of the wafer at a predetermined distance from the center of the wafer as the wafer rotates to generate displacement data (1004). The predetermined distance may correspond to a location at which the displacement sensor reads displacement data from the wafer. The predetermined distance may be located along an outer edge of the wafer at any point along a radial line extending from the center of the wafer, and / or at an angular displacement from an alignment sensor or from another displacement sensor, as described above in FIG. 5A to FIG. 5D The displacement data may represent displacement measurements from a plurality of different displacement sensors. The displacement sensors may be implemented using laser displacement sensors or any other type of displacement sensor configured to measure distance.

[0079] The method may additionally include using the displacement data to control a polishing or cleaning process performed on the wafer (1006). This portion of the method may be optional, and some implementations of the method may use the displacement data to characterize the wafer without using the displacement data to control any other process. For example, the displacement data may be used to characterize the surface of the wafer and / or characterize or test a previous process performed on the wafer. Controlling the polishing process may include using the displacement data to adjust handling controls and / or change the orientation of the carrier head holding the wafer. Controlling the polishing process may also include using the displacement data to apply varying amounts of pressure on the back side of the wafer to compensate for bending in the wafer when the wafer is pushed against the polishing pad. Some embodiments may also use the displacement data to control other processes, such as a cleaning process.

[0080] The optional step of using displacement data to control the polishing or cleaning process performed on the wafer can be performed by the computer system of the corresponding polishing machine or cleaning machine. For example, software instructions on the computer system of the polishing machine can cause one or more processors to receive displacement data generated using a displacement sensor. As described above, the displacement data can indicate at least one area of ​​bending or warping on the wafer. These instructions can also cause the machine to identify the parameters of the carrier head corresponding to the position of the bending or warping on the area of ​​the wafer. The computer system can then adjust the parameters on the carrier head to compensate for the bending or warping area on the wafer. As described above, these parameters can adjust the angle at which the carrier head grabs the wafer, the pressure applied during the polishing process, the orientation of the wafer during the process, the pressure applied to the back of the wafer using each chamber in the carrier head, etc.

[0081] It should be understood that Fig.10 The specific steps shown in the embodiment provide specific methods for measuring the displacement of wafers between semiconductor manufacturing processes according to various embodiments. Other sequences of steps may also be performed according to alternative embodiments. For example, alternative embodiments may perform the steps outlined above in a different order. In addition, Fig.10 The independent steps shown in the figure may include multiple sub-steps, which may be suitable for each sequence of independent steps. In addition, additional steps may be added or removed depending on the specific application. Many variations, modifications, and substitutions also fall within the scope of the present disclosure. Machine learning control using displacement data

[0082] As described above, the process of generating metrology data for a wafer can be a computationally complex and time consuming process, often requiring special metrology chambers that can impose significant costs on the wafer manufacturer. For example, a metrology station may perform a laser scan of the surface of a wafer to identify displacements, such as bending or warping of the wafer shape, as well as irregularities in the surface or shape of the wafer. Metrology data can also be used to measure the thickness of thin films and other features on the wafer. For example, metrology data may include the thickness or location of metal traces, semiconductor circuits, or other features that have been formed on the wafer. A metrology station that performs a laser scan or other analysis of a wafer may require a significant amount of time to perform this analysis, which may take from several minutes to as much as an hour, depending on the size of the wafer and the resolution of the scan.

[0083] Some embodiments described herein may use displacement data generated using the systems and methods described above to replace and / or generate metrology data without the need to use a metrology station for each new wafer. This practical application of machine learning principles using specific displacement data inputs to generate control signals or metrology data significantly improves the wafer manufacturing process by eliminating the need for complex metrology stations and the need to perform complex metrology scans for each new wafer. Alternatively, the displacement sensor system described above may be used to generate displacement data, and the trained model may then be used to generate the necessary control signals for polishing or cleaning processes. Some embodiments may also generate or select metrology data that may represent a wafer without performing a complete metrology scan. This increases the throughput of the wafer manufacturing process, and metrology stations may be eliminated from the wafer manufacturing pipeline. This in turn reduces abrasion and wear on the metrology station, and increases its service life.

[0084] Fig.11 A block diagram 1100 of a process for training a model according to some embodiments is shown. As described above, one or more displacement sensors may be positioned relative to a rotating platform to receive a wafer 1104. As the platform rotates, the one or more displacement sensors may measure displacement data 1112 for the wafer 1104. The one or more displacement sensors may be part of a displacement measurement system 1106, which may include any of the embodiments described above. For example, the displacement measurement system 1106 may include an alignment sensor that locates a notch in the wafer 1104. Typically, the displacement sensor measures the displacement of the wafer 1104 relative to the one or more displacement sensors. For example, the one or more displacement sensors may measure the distance between the one or more displacement sensors and the surface of the wafer 1104 as the wafer 1104 rotates.

[0085] In addition to generating displacement data 1112, other sensors 1108 may be present in some embodiments to generate additional sensor data. Sensor 1108 may be part of displacement measurement system 1106. Sensor 1108 may alternatively or additionally be included in other systems or processing chambers used to process wafer 1104 earlier in the wafer manufacturing pipeline. Sensor 1108 may generate sensor data 1114, which may also be used as input when training or using model 1122. Sensor 1108 may include a camera that captures an image of wafer 1104 as sensor data 1114. Sensor 1108 may also include an eddy current sensor that may be used to determine the thickness of a copper film on wafer 1104. Sensor 1108 may also include a rotational accelerometer or other sensor configured to measure the position or geometry of wafer 1104. Note that some embodiments may use displacement data 1112 alone, while other embodiments may augment displacement data 1112 with additional sensor data 1114. Other embodiments may rely on sensor data 1114 without requiring displacement data 1112 .

[0086] Model 1122 may be trained based on the type of output that model 1122 needs to generate. Fig.11 In an example of , the model 1122 may be trained to generate metrology data. To train the model 1122, the metrology station 1102 may perform metrology analysis on the wafer 1104. The metrology station 1102 may generate metrology data 1110. In some embodiments, a complete set of metrology data from the metrology station 1102 may be used as the metrology data 1110 for training the model 1122. In other embodiments, a subset of data points in the metrology data generated by the metrology station 1102 may be included in the metrology data 1110 for training the model 1122. For example, the metrology data 1110 may include film thickness or displacement measurements at a subset of predetermined locations on the wafer 1104. The displacement data for the wafer 1104 may be captured at multiple points in a radius across the surface of the wafer 1104. Note that the displacement data 1112 from the displacement measurement system 1106 may include displacement data captured at a predetermined radius of the wafer 1104 as the wafer 1104 rotates. The metrology data 1110 may include an entire 2D set of measurements across the surface of the wafer 1104 , which may include a predetermined radius of the displacement measurement system 1106 , along with displacement measurements at multiple additional radii to map the entire surface of the wafer 1104 .

[0087] To train the model 1122, the displacement data 1112 along with the sensor data 1114 (optionally included) may be provided as training data 1118 for training the model 1122. The model 1122 may be implemented using any type of classification model. For example, the training data 1118 may be used by a minimization regression algorithm to set weights or node values ​​in the model 1122. For example, the model 1122 may use a logistic regression model, a k-nearest neighbor model, a decision tree, a support vector machine, The method 1110 may be implemented using a Bayesian model, a random forest, a gradient boosting model, and / or any other classification model type. The labels 1120 for the training data 1118 may be generated using the metrology data 1110. For example, the model 1122 may be configured to output a classification of the shape of the wafer 1104. The metrology data 1110 may be used to select one of the shape outputs of the model 1122 to use as the label 1120 on the training data 1118. Alternatively, the model 1122 may be configured to output a displacement map that includes displacement measurements of locations on the wafer 1104, and the metrology data 1110 may use the displacement values ​​in the 2D map of the wafer 1104 as the label 1120 for the training data 1118.

[0088] Notice that Fig.11 The example of uses metrology data to train model 1122 to generate characteristic or metrology data for a wafer. This allows metrology data to be generated based on displacement data 1112 by model 1122 without requiring metrology station 1102 for subsequent wafers processed by the system when using trained model 1122. However, other models 1122 may be trained to generate control signals for cleaning or polishing processes rather than metrology data. Although in Fig.11 Although not explicitly shown in FIG. 1 , the metrology data 1110 used as the labels 1120 for training the model 1122 may be replaced with control signals that match the geometry of the wafer 1104 .

[0089] For example, the metrology data 1110 may be used to select or calculate control signal values ​​for a cleaning or polishing process. If a bow on one side of the wafer 1104 is detected by the displacement data 1112 or the metrology data 1110, the control signal for the polishing process may increase the polishing pressure on a subarea of ​​the polishing head behind the area of ​​the wafer 1104 having the bow. Depending on the embodiment, these control signals may be selected as flags 1120 based on the metrology data 1110 or the displacement data 1112.

[0090] Fig.12 A block diagram 1200 of a system for controlling a cleaning or polishing process using a trained model is shown in accordance with some embodiments. Fig.11After training the model 1122 with the process described in , the process may bypass the metrology station and use the displacement data 1112 instead. The displacement data 1112 may be provided as an input to the trained model 1122. Optionally, some embodiments may also include other sensor data from additional sensors if such sensor data is used to train the model 1122. The model 1122 may propagate these inputs through its internal layers and nodes of the model 1122 to generate a classification output that classifies the displacement data 1112. This classification may generate an output that may be used to control the polishing or cleaning process.

[0091] In this example, controlling the cleaning or polishing process may include selecting from a plurality of different cleaning or polishing processes. For example, the first cleaning or polishing process 1202-1 may be performed on a substantially flat wafer. Therefore, the polishing or cleaning process may be performed at a higher rate. In contrast, the second cleaning or polishing process 1202-2 may be configured to polish a wafer with a significant bend or warp, and thus the wafer may be polished more carefully at a slower rate. Therefore, the classification performed by the model 1122 may be integrated into the entire wafer manufacturing pipeline to independently classify the wafers and maintain parallel manufacturing channels optimized for the wafer type. By funneling substantially flat wafers into the first polishing or cleaning process 1202-1, these wafers may be polished significantly faster, and the bottleneck wafers may be funneled into the second polishing or cleaning process 1202-2, which may be performed in parallel and separately. This increases the total throughput of the process and allows the speed of each independent polishing or cleaning process to be optimized for a specific wafer geometry without the need to accommodate any wafer geometry. For example, rather than selecting a polishing rate that will accommodate all wafer surface geometries, each of the plurality of polishing or cleaning processes 1202 may be optimized to handle a specific type (eg, flat, curved, warped, "potato chip" shape, etc.).

[0092] Fig.13 A block diagram 1300 of a process for using a model to generate control signals for a cleaner polishing process according to some embodiments is shown. In this example, the model 1122 may have been trained to generate specific control signals that can be used to control a cleaning or polishing process of a wafer. When receiving displacement data 1112 as input, the model 1122 may generate a control signal 1302 that can be provided as part of a recipe, or a real-time control signal for a cleaning or polishing process 1202-1. For example, the output of the model 1122 may be included as part of a recipe, wherein the recipe includes control signals for all processes that are part of a semiconductor manufacturing step. The recipe may then be provided to the cleaning or polishing process to be performed. Alternatively, the control signal 1122 may be provided directly to the process when the process is running. This allows the model 1122 to provide real-time control for the cleaning or polishing process.

[0093] Despite Fig.13 Although not explicitly shown in FIG. , some embodiments may include Fig.12 to generate metrology data. For example, model 1122 may generate metrology data as described above. The metrology data may then be used to select or set Fig.13 For example, the film thickness determined from the metrology data can be used to control the polishing rate or pressure for that portion of the wafer.

[0094] The control signals generated by the model 1122 may include any of the inputs that may be provided to any semiconductor manufacturing process. In this specific case of a polishing or cleaning process, a control signal may be provided to control the rotation speed 1302-1 of a platform or pedestal that supports a wafer during the cleaning or polishing process. The control signal may also include a head rotation speed 1302-2 for a head that rotates the wafer and holds the wafer against a polishing or cleaning pad. As described above, the rotation speed for each of these two parameters may be adjusted based on displacement data and the surface geometry of the wafer. The control signal may also control the flow of the slurry used in the polishing or cleaning process. This may include the flow rate of the slurry, as well as the type of slurry used. For example, when more aggressive polishing is required to improve the total processing time, a more abrasive slurry may be used. As described above, some polishing or cleaning heads may include partitions that allow different pressures to be applied to different portions of the wafer in the polishing or cleaning head. The control signal may include a partition pressure 1302-4 for each of these different partitions. For example, increasing the pressure on a specific partition corresponding to a curved portion of the wafer may provide more aggressive polishing or cleaning to this wafer area.

[0095] Note that in Fig.13 The control signals provided to the cleaning or polishing process 1202-1 shown in the figure are provided by way of example only and are not intended to be limiting. The cleaning or polishing process may include many additional controls, such as temperature control, flow rates for various gases or other substances in the chamber, chamber pressure control, and / or other environmental parameter controls. Some embodiments may also provide each of these controls as a time series of values ​​so that the control values ​​change over time. For example, the model 1122 may be trained to generate a time series output of the speed of the head rotation speed 1302-2. A high speed may be used first in order to flatten the curved portion of the wafer, and the speed may be gradually slowed to perform a more uniform and careful polishing of the flattened wafer as a whole.

[0096] Fig.14Block diagram 1400 illustrates how a model according to some embodiments may be used to generate metrology data 1402. Note that model 1122 is not necessarily required to provide a complete set of values ​​in metrology data 1402. Instead, some embodiments may provide a subset of values ​​in metrology data 1402. For example, displacement data 1112 and / or other sensor data 1116 may be used to characterize the thickness of a film on the surface of a wafer. This thickness may be used to populate a field in metrology data 1402. Thus, metrology data 1402 may include less than a complete set of metrology data 1402.

[0097] In some embodiments, a subset of the values ​​provided by the model 1122 may be combined with other values ​​from an actual metrology process to generate a complete metrology data set 1402. For example, a metrology process may be reduced so that only a subset of the values ​​are measured and the total metrology process time is reduced. The remaining values ​​may be filled in using the model 1122. In other embodiments, the model 1122 may estimate all of the metrology data in the metrology data 1402. For example, many of the values ​​in the metrology data 1402 may be strongly correlated with the shape and topology of the wafer measured by the displacement data 1112. The model 1122 may be trained to recognize these cross-correlations and generate estimates of the metrology data 1402 based on the measured displacements.

[0098] In some embodiments, model 1122 may be trained to select a set of pre-existing metrology data that can be used to populate metrology data 1402. For example, metrology data 1402 may include data fields that characterize the thickness of a film on a wafer. Multiple sets of values ​​for these thickness data fields may be stored from previous wafer testing or training data. When model 1122 receives sensor data 1116, model 1122 may be configured to select one of the pre-existing data sets that characterize the thickness of a film on a wafer. The same process may be used to select a pre-existing data set for any other set of fields in metrology data 1402.

[0099] Fig.15 A flow chart 1500 of a method for training and using a model with displacement data according to some embodiments is shown. The method may include receiving displacement data generated using a displacement sensor (1502). The displacement data may be measured for a wafer, and the displacement data may indicate a displacement of the wafer relative to the displacement sensor. In some embodiments, this may include rotating the wafer about a center of the wafer and measuring a displacement of the wafer at a predetermined distance from the center of the wafer. Some embodiments may also measure additional sensor data as described above.

[0100] The method may additionally include providing the displacement data to a model trained using the metrology data set and the corresponding displacement data set (1504). For example, the displacement data may be provided to the model as a labeled training data set from a previous wafer process. The model may be trained as a classification model to recognize known patterns and other inputs in the displacement data and then select a corresponding set of outputs based on the classification of the inputs.

[0101] The method may further include generating metrology data for the wafer based on output from the model in response to providing the displacement data as input to the model (1506). Some embodiments may generate control signals for a cleaning or polishing chamber based on the output from the model. The output of the model may be used to control a cleaning or polishing process for the same wafer in a subsequent processing chamber. For example, multiple cleaning or polishing lines with varying speeds may be arranged, where the output of the model routes the wafer to different polishing processes based on the flatness of the wafer. The output from the model may include control signals for controlling polishing pressure, speed, slurry flow, polishing material, etc. based on the wafer displacement data.

[0102] Each of the methods described herein can be implemented by a computer system. Each step of these methods can be automatically performed by a computer system, and / or can be provided with input / output involving a user. For example, a user can provide input for each step in the method, and each of these inputs can be a specific output in response to requesting such input, wherein the output is generated by a computer system. Each input can be received in response to a corresponding request output. In addition, the input can be received from a user, received as a data stream from another computer system, retrieved from a memory location, retrieved on a network, requested from a web service, etc. Similarly, the output can be provided to a user, another computer system as a data stream, stored in a memory location, sent on a network, provided to a web service, etc. In short, each step of the method described herein can be performed by a computer system, and can involve any number of inputs, outputs, and / or requests to and from a computer system that may or may not involve a user. Those steps that do not involve a user can be considered to be automatically performed by a computer system without human intervention. Thus, it will be understood that, in light of the present disclosure, each step of each method described herein may be modified to include input and output to and from a user, or may be performed automatically by a computer system without human intervention, with any determinations made by a processor. Additionally, some embodiments of each of the methods described herein may be implemented as a set of instructions stored on a tangible, non-transitory storage medium to form a tangible software product.

[0103] Fig.16An exemplary computer system 1600 is shown in which various embodiments may be implemented. The computer system 1600 may be used to implement any of the computer systems described above. For example, the computer system 1600 may be used to control a displacement sensor and store displacement data. The computer system 1600 may also be used to send displacement data to other semiconductor manufacturing machines, such as a CMP machine or a cleaner. The CMP machine and / or cleaner may also include a computer system 1600 that uses the displacement data to control a polishing and / or cleaning process.

[0104] As shown, computer system 1600 includes a processing unit 1604 that communicates with a number of peripheral subsystems via a bus subsystem 1602. These peripheral subsystems may include a processing acceleration unit 1606, an I / O subsystem 1608, a storage subsystem 1618, and a communication subsystem 1624. Storage subsystem 1618 includes tangible computer-readable storage media 1622 and system memory 1610.

[0105] The bus subsystem 1602 provides a mechanism for making the various components and subsystems of the computer system 1600 communicate with each other according to the intention. Although the bus subsystem 1602 is schematically shown as a single bus, alternative embodiments of the bus subsystem can utilize multiple buses. The bus subsystem 1602 can be any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, or a local bus using any of various bus architectures. For example, such an architecture can include an industrial standard architecture (ISA) bus, a microchannel architecture (MCA) bus, an enhanced ISA (EISA) bus, a video electronics standard association (VESA) local bus, and a peripheral component interconnect (PCI) bus, which can be implemented as a Mezzanine bus manufactured according to the IEEE P1386.1 standard.

[0106] The processing unit 1604, which may be implemented as one or more integrated circuits (e.g., conventional microprocessors or microcontrollers), controls the operation of the computer system 1600. One or more processors may be included in the processing unit 1604. These processors may include single-core or multi-core processors. In some embodiments, the processing unit 1604 may be implemented as one or more independent processing units 1632 and / or 1634, wherein a single-core or multi-core processor is included in each processing unit. In other embodiments, the processing unit 1604 may also be implemented as a quad-core processing unit formed by integrating two dual-core processors into a single chip.

[0107] In various embodiments, the processing unit 1604 can execute various programs in response to program code, and can maintain multiple concurrently executed programs or processes. At any given time, some or all of the program code to be executed may reside in (multiple) processors 1604 and / or in the storage subsystem 1618. Through appropriate programming, the processor 1604 can provide the various functionalities described above. The computer system 1600 may additionally include a processing acceleration unit 1606, which may include a digital signal processor (DSP), a special processor, and the like.

[0108] I / O subsystem 1608 may include user interface input devices and user interface output devices. User interface input devices may include a keyboard, a pointing device (such as a mouse or trackball), a touch pad or touch screen integrated into a display, a scroll wheel, a click wheel, a dial, buttons, switches, a keypad, an audio input device with a voice command recognition system, a microphone, and other types of input devices. User interface input devices may include, for example, a Microsoft Motion sensing and / or gesture recognition devices such as motion sensors that enable a user to control and interact with input devices through a natural user interface using gestures and verbal commands, such as Microsoft 360 game controller. User interface input devices may also include such as Google Eye gesture recognition devices such as blink detectors that detect eye activity from a user (e.g., "blinking" when taking a picture and / or making a menu selection) and translate the eye gesture into an input device (e.g., Google ) input. In addition, the user interface input device may include a user to enable the user to communicate with the voice recognition system (e.g., Navigator) interactive voice recognition sensing device.

[0109] The user interface input device may also include, but is not limited to, a three-dimensional (3D) mouse, a joystick or indicator stick, a game keyboard and a drawing tablet, and audio / visual equipment, such as a speaker, a digital camera, a digital portable camera, a portable media player, a webcam, an image scanner, a fingerprint scanner, a barcode reader 3D scanner, a 3D printer, a laser rangefinder, and an eye gaze tracking device. In addition, the user interface input device may include, for example, a medical imaging input device, such as computed tomography, magnetic resonance imaging, position emission tomography, medical ultrasound examination equipment. The user interface input device may also include, for example, an audio input device, such as a MIDI keyboard, a digital musical instrument, etc.

[0110] The user interface output device may include a display subsystem, indicator lights, or non-visual displays such as audio output devices, etc. The display subsystem may be a cathode ray tube (CRT), a flat panel device (such as a flat panel device using a liquid crystal display (LCD) or a plasma display), a projection device, a touch screen, etc. In general, the use of the term "output device" is intended to include all possible types of devices and mechanisms for outputting information from the computer system 1600 to a user or other computer. For example, the user interface output device may include, but is not limited to, various display devices that visually convey text, graphics, and audio / video information, such as cameras, printers, speakers, headphones, automatic navigation systems, plotters, voice output devices, and modems.

[0111] Computer system 1600 may include storage subsystem 1618, including software elements shown as currently located in system memory 1610. System memory 1610 may store program instructions loadable and executable on processing unit 1604, as well as data generated during the execution of these programs.

[0112] Depending on the configuration and type of computer system 1600, system memory 1610 may be volatile (such as random access memory (RAM)) and / or non-volatile (such as read-only memory (ROM), flash memory, etc.). RAM typically contains data and / or program modules that are immediately accessible to and / or currently being operated and executed by processing unit 1604. In some embodiments, system memory 1610 may include a variety of different types of memory, such as static random access memory (SRAM) or dynamic random access memory (DRAM). In some embodiments, a basic input / output system (BIOS) containing basic routines such as those that help transfer information between elements within computer system 1600 during startup may typically be stored in ROM. By way of example, and not limitation, system memory 1610 also illustrates application programs 1612 (which may include client applications, web browsers, mid-tier applications, relational database management systems (RDBMS), etc.), program data 1614, and operating system 1616. For example, operating system 1616 may include various versions of Microsoft apple and / or Linux operating systems, various commercially available or UNIX-like operating systems (including but not limited to various GNU / Linux operating systems, Google OS, etc.) and / or mobile operating systems such as iOS, Phone, OS, 10OS, and OS operating system.

[0113] Storage subsystem 1618 may also provide a tangible computer-readable storage medium for storing basic programming and data constructs that provide the functionality of some embodiments. Software (programs, code modules, instructions) that provide the functionality described above when executed by a processor may be stored in storage subsystem 1618. These software modules or instructions may be executed by processing unit 1604. Storage subsystem 1618 may also provide a repository for storing data used in accordance with some embodiments.

[0114] Storage subsystem 1600 may also include a computer-readable storage media reader 1620, which may be further connected to computer-readable storage media 1622. Together with and optionally in conjunction with system memory 1610, computer-readable storage media 1622 may collectively represent remote, local, fixed, and / or removable storage devices plus storage media for transiently and / or more permanently containing, storing, transmitting, and retrieving computer-readable information.

[0115] The computer-readable storage medium 1622 containing the code, or portions of the code, may also include any suitable media, including storage media and communication media, such as, but not limited to, volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing and / or transmitting information. This may include tangible computer-readable storage media, such as RAM, ROM, electronically erasable programmable ROM (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disk (DVD), or other optical storage, cassettes, magnetic tape, disk storage or other magnetic storage devices, or other tangible computer-readable media. This may also include non-tangible computer-readable media, such as data signals, data transmissions, or any other media that can be used to send the desired information and can be accessed by the computing system 1600.

[0116] By way of example, computer-readable storage media 1622 may include a hard disk drive that reads from or writes to a non-removable non-volatile magnetic media, a magnetic disk drive that reads from or writes to a removable non-volatile magnetic disk, and a magnetic disk drive that reads from or writes to a removable non-volatile optical disk (such as a CD ROM, a DVD, a CD-ROM, or other optical media) to read or write removable non-volatile optical disks. Computer-readable storage media 1622 may include, but is not limited to The computer readable storage medium 1622 may include a solid-state drive (SSD) based on nonvolatile memory, such as a flash memory-based SSD, an enterprise flash drive, a solid-state ROM, etc.; a volatile memory-based SSD, such as a solid-state RAM, a dynamic RAM, a static RAM, a DRAM-based SSD, a magnetoresistive RAM (MRAM) SSD; and a hybrid SSD using a combination of DRAM and a flash memory-based SSD. The disk drive and its associated computer-readable media may provide nonvolatile storage of computer-readable instructions, data structures, program modules, and other data for the computer system 1600.

[0117] The communication subsystem 1624 provides an interface to other computer systems and networks. The communication subsystem 1624 is used as an interface to receive data from other systems from the computer system 1600 and to send data to other systems. For example, the communication subsystem 1624 can enable the computer system 1600 to be connected to one or more devices via the Internet. In some embodiments, the communication subsystem 1624 may include a radio frequency (RF) transceiver component for accessing a wireless voice and / or data network (e.g., using cellular phone technology, advanced data network technology, such as 3G, 4G or EDGE (Enhanced Data Rates for Global Evolution)), WiFi (IEEE 802.11 family standards, or other mobile communication technologies, or any combination thereof), a global positioning system (GPS) receiver component, and / or other components. In some embodiments, as a supplement or alternative to a wireless interface, the communication subsystem 1624 can provide wired network connectivity (e.g., Ethernet).

[0118] In some embodiments, communication subsystem 1624 may also receive incoming communications in the form of structured and / or unstructured data feeds 1626 , event streams 1628 , event updates 1630 , and the like on behalf of one or more users of computer system 1600 .

[0119] For example, the communication subsystem 1624 may be configured to receive data feeds 1626 in real time from users of social networks and / or other communication services, such as feed, Updates, web page feeds (such as Rich Site Summary (RSS) feeds), and / or real-time updates from one or more third-party information sources.

[0120] Additionally, the communication subsystem 1624 may also be configured to receive data in the form of a continuous data stream, which may include an event stream 1628 of real-time events and / or event updates 1630 that may be continuous or unbounded in nature without a clear end. Examples of applications that generate continuous data may include, for example, sensor data applications, financial instruments, network performance measurement tools (e.g., network monitoring and traffic management applications), clickstream analysis tools, automated traffic monitoring, etc.

[0121] The communication subsystem 1624 may also be configured to output structured and / or unstructured data feeds 1626 , event streams 1628 , event updates 1630 , etc. to one or more databases that may be in computer communication with one or more streaming data sources coupled to the computer system 1600 .

[0122] Computer system 1600 may be one of various types, including a handheld portable device (e.g., Cellular phone, computing tablets, PDAs), wearable devices (e.g., Google head mounted display), PC, workstation, mainframe, kiosk, server rack, or any other data processing system.

[0123] Due to the ever-changing nature of computers and networks, the description of the computer system 1600 depicted in the drawings is intended only as a specific example. Many other configurations with more or fewer components than the system depicted in the drawings are possible. For example, custom hardware may also be used and / or specific elements may be implemented in hardware, firmware, software (including small applications), or a combination. In addition, connections to other computing devices (such as network input / output devices) may be employed. Based on the disclosure and teachings provided herein, other ways and / or methods of implementing various embodiments should be apparent.

[0124] In the above description, for the purpose of explanation, several specific details are set forth in order to provide a thorough understanding of various embodiments. However, it will be apparent that some embodiments may be practiced without some of these specific details. In other examples, well-known structures and devices are shown in the form of block diagrams.

[0125] The above description provides only exemplary embodiments and is not intended to limit the scope, applicability, or configuration of the present disclosure. On the contrary, the above description of each embodiment will provide an enabling disclosure for implementing at least one embodiment. It should be understood that various changes in the function and arrangement of elements may be made without departing from the spirit and scope of some embodiments as set forth in the appended claims.

[0126] Specific details are given in the above description to provide a thorough understanding of the present disclosure. However, it will be understood that the embodiments may be practiced without these specific details. For example, circuits, systems, networks, processes, and other components may be shown as components in the form of block diagrams so as not to obfuscate the embodiments with unnecessary details. In other examples, well-known circuits, processes, algorithms, structures, and techniques may be shown without unnecessary details so as to avoid obfuscating the embodiments.

[0127] In addition, it is noted that various embodiments may be described as processes, which are depicted as flow charts, flow diagrams, data flow diagrams, structure diagrams, or block diagrams. Although a flow chart may describe an operation as a sequential process, many operations may be performed in parallel or simultaneously. In addition, the order of the operations may be rearranged. A process terminates when its operations are completed, but may have additional steps not included in the accompanying drawings. A process may correspond to a method, function, program, subroutine, subprogram, etc. When a process corresponds to a function, its termination may correspond to the function returning to a calling function or a main function.

[0128] The term "computer-readable medium" includes, but is not limited to, portable or fixed storage devices, optical storage devices, wireless channels, and various other media capable of storing, containing, or carrying instructions and / or data. Code segments or machine executable instructions can represent any combination of programs, functions, subroutines, programs, routines, subroutines, modules, software encapsulation, categories, or instructions, data structures, or program statements. Code segments can be coupled to another code segment or hardware circuit by transmitting and / or receiving information, data, independent variables, parameters, or memory contents. Information, independent variables, parameters, data, etc. can be transmitted, forwarded, or sent via any suitable component, including memory sharing, message passing, token passing, network transmission, etc.

[0129] In addition, the embodiments may be implemented by hardware, software, firmware, middleware, microcode, hardware description language, or any combination thereof. When implemented in software, firmware, middleware, or microcode, program code or code segments for performing the necessary tasks may be stored in a machine-readable medium. (Multiple) processors may perform the necessary tasks.

[0130] In the above description, features are described with reference to specific embodiments thereof, but it should be appreciated that not all embodiments are limited thereto. The various features and aspects of some embodiments may be used independently or in combination. In addition, the embodiments may be utilized in any number of environments and applications beyond those described herein without departing from the broad spirit and scope of the specification. The specification and drawings are therefore to be considered illustrative rather than restrictive.

[0131] In addition, for the purpose of illustration, the method is described in a specific order. It should be appreciated that in an alternative embodiment, the method can be performed in a different order than described. It should also be appreciated that the method described above can be performed by hardware components or can be embodied in a sequence of machine executable instructions, which can be used to make a machine (such as a general or special processor or logic circuit) programmed with instructions perform the method. These machine executable instructions can be stored on one or more machine-readable media, such as CD-ROM or other types of optical disks, floppy disks, ROM, RAM, EPROM, EEPROM, magnetic cards or optical cards, flash memory, or other types of machine-readable media suitable for storing electronic instructions. Alternatively, the method can be performed by a combination of hardware and software.

Claims

1. A system comprising: one or more displacement sensors positioned to measure displacement of the wafer as the wafer rotates on the platform; a cleaning or polishing chamber configured to clean or polish the wafer; as well as A computer system configured to perform operations including: receiving, for the wafer, displacement data generated using the one or more displacement sensors, wherein the displacement data indicates a displacement of the wafer relative to the one or more displacement sensors; providing the displacement data to a model trained using a displacement dataset; generating a control signal for the cleaning or polishing chamber based on an output from the model in response to receiving the displacement data as input; as well as The control signal is used to control the cleaning or polishing process of the wafer.

2. The system of claim 1 , wherein controlling the cleaning or polishing process of the wafer comprises selecting the cleaning or polishing process from a plurality of different cleaning or polishing processes, wherein the cleaning or polishing process has a shorter run time than at least one of the plurality of different cleaning or polishing processes.

3. The system of claim 1, wherein controlling the cleaning or polishing process of the wafer comprises controlling a rotation speed of a platform supporting the wafer during the cleaning or polishing process.

4. The system of claim 1, wherein controlling the cleaning or polishing process of the wafer comprises controlling a rotation speed of a cleaning or polishing head during the cleaning or polishing process.

5. The system of claim 1, wherein controlling the cleaning or polishing process of the wafer comprises controlling a flow of a slurry used in the cleaning or polishing process.

6. The system of claim 1, wherein controlling the cleaning or polishing process of the wafer comprises controlling pressure applied to zones in a cleaning or polishing head during the cleaning or polishing process.

7. The system of claim 1, wherein the displacement data indicates areas of bowing or warping on the wafer.

8. A method comprising: For a wafer, receiving displacement data generated using a displacement sensor, wherein the displacement data indicates a displacement of the wafer relative to the displacement sensor; providing the displacement data to a model trained using a metrology dataset and a corresponding displacement dataset; as well as In response to receiving the displacement data as input to the model, metrology data for the wafer is generated based on output from the model.

9. The method of claim 8, further comprising: The metrology data is used to control a cleaning or polishing process of the wafer.

10. The method of claim 8, wherein the metrology data includes the thickness of a film or feature on the wafer.

11. The method of claim 8, further comprising: rotating the wafer about the center of the wafer; as well as The displacement of the wafer at a predetermined distance from the center of the wafer is measured as the wafer rotates to generate the displacement data.

12. The method of claim 11, further comprising: The displacement of the baseline wafer is measured to generate baseline displacement data representative of a flat wafer surface.

13. The method of claim 12, wherein generating the displacement data comprises: A displacement of the wafer relative to the baseline displacement data is determined.

14. The method of claim 11, further comprising: The displacement data is mapped to two-dimensional coordinates on the wafer based on the alignment data.

15. The method of claim 11, further comprising: The displacement data is mapped to a rotation angle on the wafer based on the alignment data.

16. The method of claim 11, further comprising: A displacement region on the wafer in the displacement data is identified and the displacement region is provided to the polishing or cleaning process.

17. A non-transitory computer readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising: receiving, for a wafer, displacement data generated using one or more displacement sensors, wherein the displacement data indicates a displacement of the wafer relative to the one or more displacement sensors; providing the displacement data to a model trained using a displacement dataset; and In response to receiving the displacement data as input, a control signal for cleaning or polishing the chamber is generated based on output from the model.

18. The non-transitory computer readable medium of claim 17, wherein the operations further comprise: The control signal is used to control the cleaning or polishing process of the wafer.

19. The non-transitory computer readable medium of claim 18, wherein controlling the cleaning or polishing process of the wafer comprises controlling a rotation speed of a platform supporting the wafer during the cleaning or polishing process.

20. The non-transitory computer readable medium of claim 18, wherein controlling the cleaning or polishing process of the wafer comprises selecting the cleaning or polishing process from a plurality of different cleaning or polishing processes.

Images