Integrated adaptive positioning system and routine for automated wafer handling robot teaching and health check

Abstract

Systems and techniques for determining and using multiple types of offsets of a wafer support for providing wafers to a wafer station of a semiconductor processing tool are disclosed; such techniques and systems may use an auto-calibration wafer that may have a plurality of sensors including a plurality of edge-located imaging sensors that may be used to image fiducials associated with two different structures located in a selected wafer station.

Description

This application is a divisional application of the invention patent application with application number 202080066598.8, application date July 21, 2020, applicant is Rum Research Company, and the invention name is "Integrated adaptive positioning system and routine for teaching and health inspection of automated wafer handling robots". Incorporated by Reference

[0001] The PCT application form is filed concurrently with this specification as a part of this application. Each application to which this application claims the benefit of or priority as identified in the concurrently filed PCT application form is incorporated herein by reference in its entirety and for all purposes. Background Art

[0002] Semiconductor processing tools use wafer handling robots to move semiconductor wafers between various wafer stations. Because wafer handling robots typically utilize a handle- or spoon-shaped end effector to pick up semiconductor wafers, and because semiconductor wafers are not securely secured to the end effector of the wafer handling robot, there is often a small degree of relative placement variation between the end effector and the semiconductor wafer placed thereon. Due to the sensitive nature of semiconductor processing operations, such variations in placement of semiconductor wafers by wafer handling robots are typically corrected to ensure that the semiconductor wafers are placed within an acceptable tolerance range at a desired location within their respective processing stations, such as approximately centered within the processing station. Modern semiconductor processing tools utilize active wafer centering (AWC) systems to assist in such wafer placement. Summary of the Invention

[0003] Details of one or more embodiments of the subject matter described in this specification are set forth in the accompanying drawings and the following description. Other features, aspects, and advantages will become apparent from the description, drawings, and claims. The following non-limiting implementations should be considered part of this disclosure; other implementations will be apparent from the disclosure and the drawings as a whole.

[0004] In some implementations, a system for facilitating calibration of a wafer handling robot of a semiconductor processing tool can be provided. The system can include, for example, an automated calibration wafer comprising: a substrate sized to be transported by the wafer handling robot and having a first side configured to contact an end effector of the wafer handling robot when the substrate is being transported by the wafer handling robot; a plurality of first imaging sensors supported by the substrate and positioned at a plurality of locations offset from a common point on the substrate, each first imaging sensor having a downwardly facing field of view when the substrate is positioned with the first side facing downward; and a first controller communicatively coupled to each of the first imaging sensors.

[0005] In some implementations of the system, the first imaging sensors may be arranged in a circular array around the common point.

[0006] In some implementations of the system, the substrate can be nominally circular and can have the same diameter as a semiconductor wafer that the semiconductor processing tool is configured to process.

[0007] In some implementations of the system, the substrate can be nominally circular and can have the same diameter as an edge ring with which the semiconductor processing tool is configured to be used.

[0008] In some implementations of the system, the substrate can be nominally circular and can have a diameter between an outer diameter and an inner diameter of an edge ring with which the semiconductor processing tool is configured to be used.

[0009] In some implementations of the system, the substrate can be nominally circular and can have a diameter within ±10% of a mean between an outer diameter and an inner diameter of an edge ring with which the semiconductor processing tool is configured to use the edge ring.

[0010] In some implementations of the system, the substrate can be a nominal circular disk, and the circular disk can have a diameter selected from the group consisting of 200 mm, 300 mm, and 450 mm.

[0011] In some implementations of the system, the auto-calibration wafer may further include a power supply configured to provide power to at least the first controller and the first imaging sensor.

[0012] In some implementations of the system, the power source may be a rechargeable battery, and the auto-calibration chip may further include a wireless charging feature configured to charge the rechargeable battery when the rechargeable battery is interfaced with an electromagnetic field.

[0013] In some implementations of the system, the automatic calibration chip may further include a first wireless communication interface, and the first wireless communication interface is communicatively connected to the first controller.

[0014] In some implementations of the system, the first wireless communication interface may include one or more wireless communication interfaces, such as a Bluetooth transceiver or a WiFi transceiver.

[0015] In some implementations of the system, the auto-calibration wafer may further include one or more position sensors, and the one or more position sensors may be communicatively coupled to the first controller.

[0016] In some implementations of the system, each of the orientation sensors can be an inclinometer or an accelerometer.

[0017] In some implementations of the system, the automatic calibration wafer may further include one or more vibration sensors, and the one or more vibration sensors may be communicatively coupled to the first controller.

[0018] In some implementations of the system, each shock sensor can be an accelerometer, a laser microphone, or an optical distance measurement sensor.

[0019] In some implementations of the system, the automatic calibration chip may also include one or more proximity sensors, each proximity sensor being configured to measure a distance between the first side and an object located below the proximity sensor when the first side is facing downward, and the one or more proximity sensors may be communicatively connected to the first controller.

[0020] In some implementations of the system, each proximity sensor can be an optical proximity sensor, an inductive proximity sensor, or a capacitive proximity sensor.

[0021] In some implementations of the system, the first imaging sensors may be arranged in a circular array around the common point, the substrate may be nominally circular and may have the same diameter as the semiconductor wafer that the semiconductor processing tool is configured to process, the substrate may be a nominal disc and the disc has a diameter selected from the group consisting of 200 mm, 300 mm, and 450 mm, and the auto-calibration wafer may further include: a rechargeable battery configured to provide power to at least the first controller and the first imaging sensor; a wireless charging feature configured to charge the rechargeable battery when the rechargeable battery is docked with an electromagnetic field; a first wireless communication interface that may be communicatively coupled to the first controller and may include one or more wireless communication interfaces such as a Bluetooth transceiver or a WiFi transceiver; one or more vibration sensors that may be communicatively coupled to the first controller; and one or more proximity sensors, each proximity sensor being communicatively coupled to the first controller and configured to measure a distance between the first side and an object located below the proximity sensor when the first side is facing downward.

[0022] In some implementations of the system, the system may further include the semiconductor processing tool, and the semiconductor processing tool may include: a wafer handling robot; one or more wafer stations; and a second controller. In such an implementation, each wafer station may include one or more corresponding wafer supports, the wafer handling robot may be communicatively coupled to the second controller, and the second controller and the first controller may be configured to: a) select a first wafer support of the one or more wafer supports of a first wafer station of the one or more wafer stations; b) cause the wafer handling robot to position the autocalibration wafer over the first wafer station; and c) cause each first imaging sensor to obtain a corresponding first image of a fiducial of the first wafer support while the autocalibration wafer is positioned over the first wafer support.

[0023] In some implementations of the system, the second controller and the first controller may be further configured to determine position information of a center point of the first wafer support based on the first image.

[0024] In some such implementations of the system, the second controller and the first controller may also be jointly configured to: d) cause the wafer handling robot to retrieve the calibration wafer; and e) cause the wafer handling robot to transfer the calibration wafer to the first wafer support such that a center point of the calibration wafer is nominally centered on a center point of the first wafer support when viewed along a vertical axis.

[0025] In some implementation schemes of the system, the second controller and the first controller can also be configured together to: f) cause the chip handling robot to position the automatic calibration chip above the first chip support and the calibration chip; g) cause each first camera sensor to obtain a second image corresponding to the benchmark of the first chip support and the benchmark of the calibration chip when the automatic calibration chip is positioned above the first chip support and the calibration chip; and h) determine the horizontal offset of the chip / chip support between the center point of the calibration chip and the center point of the first chip support based on the gap size between the first chip support and the benchmark of the calibration chip in the second image.

[0026] In some implementations of the system, the second controller and the first controller may also be jointly configured to: i) compare the horizontal offset of the wafer / wafer support with a wafer / wafer support horizontal offset threshold; and j) in response to determining that the horizontal offset of the wafer / wafer support is greater than the wafer / wafer support horizontal offset threshold, cause the wafer handling robot to reposition the calibration wafer relative to the first wafer support to reduce the horizontal offset of the wafer / wafer support.

[0027] In some implementations of the system, the second controller and the first controller may also be jointly configured to repeat (f) to (j) N times or until the horizontal offset of the wafer / wafer support is at or below a horizontal offset threshold of the wafer / wafer support, whichever occurs first.

[0028] In some implementations of the system, the second controller and the first controller may also be collectively configured to: d) cause the wafer handling robot to retrieve a first edge ring; and e) cause the wafer handling robot to transfer the first edge ring to the first wafer support such that a center point of the first edge ring is nominally centered on a center point of the first wafer support when viewed along a vertical axis.

[0029] In some implementation schemes of the system, the second controller and the first controller can also be configured together to: f) cause the wafer handling robot to place the automatic calibration wafer above the first wafer support and the first edge ring; g) cause each first camera sensor to obtain a second image corresponding to the benchmark of the first wafer support and the benchmark of the first edge ring when the automatic calibration wafer is placed above the first wafer support and the first edge ring; and h) determine the horizontal offset of the edge ring / wafer support between the center point of the first edge ring and the center point of the first wafer support based on the gap size between the benchmark of the first wafer support and the benchmark of the first edge ring in the second image.

[0030] In some implementations of the system, the second controller and the first controller may be further jointly configured to: i) compare the horizontal offset of the edge ring / wafer support with a horizontal offset threshold of the edge ring / wafer support; and j) in response to determining that the horizontal offset of the edge ring / wafer support exceeds the horizontal offset threshold of the edge ring / wafer support, cause the wafer handling robot to reposition the first edge ring relative to the first wafer support to reduce the horizontal offset of the edge ring / wafer support.

[0031] In some implementations of the system, the second controller and the first controller may also be collectively configured to repeat (f) to (j) N times or until the horizontal offset of the edge ring / wafer support is at or below a horizontal offset threshold of the edge ring / wafer support, whichever occurs first.

[0032] In some implementations of the system, the second controller and the first controller can also be jointly configured to: f) cause the wafer handling robot to retrieve a calibration wafer; and g) cause the wafer handling robot to transfer the calibration wafer to the first wafer support so that a center point of the calibration wafer is nominally centered on the center point of the first edge ring when viewed along a vertical axis.

[0033] In some implementation schemes of the system, the second controller and the first controller can also be jointly configured to: h) cause the wafer handling robot to position the automatic calibration wafer above the first wafer support, the first edge ring and the calibration wafer; i) cause each first camera sensor to obtain a second image corresponding to the benchmark of the calibration wafer and the benchmark of the first edge ring when the automatic calibration wafer is positioned above the first wafer support, the calibration wafer and the first edge ring; and j) determine the horizontal offset of the edge ring / wafer between the center point of the first edge ring and the center point of the calibration wafer based on the gap size between the benchmark of the calibration wafer and the benchmark of the first edge ring in the second image.

[0034] In some implementations of the system, the second controller and the first controller may be further jointly configured to: k) compare the horizontal offset of the edge ring / wafer to a horizontal offset threshold of the edge ring / wafer; and l) in response to determining that the horizontal offset of the edge ring / wafer is greater than the horizontal offset threshold of the edge ring / wafer, cause the wafer handling robot to reposition the calibration wafer relative to the first edge ring to reduce the horizontal offset of the edge ring / wafer.

[0035] In some implementations of the system, the second controller and the first controller may also be collectively configured to repeat (h) to (1) M times or until the horizontal offset of the edge ring / wafer is at or below the horizontal offset threshold of the edge ring / wafer, whichever occurs first.

[0036] In some implementation schemes of the system, the second controller and the first controller can also be configured together to: cause the chip handling robot to reposition the automatic calibration chip above the first chip support, the first edge ring and the calibration chip; cause each first camera sensor to obtain a third image corresponding to the benchmark of the calibration chip and the benchmark of the first chip support when the automatic calibration chip is positioned above the first chip support, the calibration chip and the first edge ring; and determine the horizontal offset of the chip support / chip between the center point of the calibration chip and the center point of the first chip support based on the gap size between the benchmark of the first chip support and the benchmark of the calibration chip in the third image.

[0037] In some implementations of the system, the second controller and the first controller may be jointly configured to: compare the horizontal offset of the wafer support / wafer with a wafer support / wafer horizontal offset threshold; and in response to determining that the horizontal offset of the wafer support / wafer is greater than the wafer support / wafer horizontal offset threshold, cause the wafer handling robot to reposition at least one object relative to the first wafer support, the at least one object being selected from the group consisting of the calibration wafer and the edge ring.

[0038] In some implementations of the system, the semiconductor processing tool can include a semiconductor processing chamber, the first wafer station can be located in the semiconductor processing chamber, and the first wafer support can include a susceptor in the semiconductor processing chamber.

[0039] In some implementations of the system, the semiconductor processing tool may include a load lock for transferring wafers between different pressure environments, the first wafer station is located in the load lock, and the first wafer support is a structure in the load lock.

[0040] In some implementations of the system, the semiconductor processing tool may include a buffer for storing one or more wafers before, after, or between processing operations, the first wafer station may be located in the buffer, and the first wafer support may be one of a plurality of wafer support racks in the buffer.

[0041] In some implementations of the system, the semiconductor processing tool can include a load lock for transferring wafers between different pressure environments, the first wafer station can be located in the load lock, and the first wafer support can be a structure in the load lock.

[0042] In some implementations of the system, the system may further include the semiconductor processing tool, wherein the semiconductor processing tool may include: a wafer handling robot; one or more wafer stations; and a second controller. In such a system, each wafer station may include one or more corresponding wafer supports, the wafer handling robot and the second controller may be communicatively coupled, and the second controller and the first controller may be jointly configured to: a) select a first wafer support of the one or more wafer supports of a first wafer station of the one or more wafer stations; b) cause the wafer handling robot to transfer the autocalibration wafer to the first wafer station; and c) cause the one or more orientation sensors to obtain a tilt measurement of the substrate.

[0043] In some implementations of the system, the second controller can be configured to remove an edge ring from the first wafer support before performing (b).

[0044] In some implementations of the system, the system may further include the semiconductor processing tool, the semiconductor processing tool may include: a wafer handling robot; one or more wafer stations; and a second controller. In such an implementation, each wafer station may include one or more corresponding wafer supports, the wafer handling robot and the second controller may be communicatively coupled, and the second controller and the first controller may be jointly configured to: a) select a first wafer support of the one or more wafer supports of a first wafer station of the one or more wafer stations; b) cause a plurality of lift pins of the first wafer support to relative position with the first wafer support so that the lift pins protrude from the first wafer support; c) cause the wafer handling robot to transfer the auto-calibration wafer to the lift pins; d) cause the lift pins to further relative position with the first wafer support while the auto-calibration wafer is supported by the lift pins; e) obtain vibration data from the one or more vibration sensors during (d); f) evaluate the vibration data to determine whether the vibration data indicates vibration exceeding a predetermined threshold; and g) provide a notification when the vibration data exceeds the predetermined threshold.

[0045] In some implementations of the system, as part of (d), the second controller can be configured to cause a further relative displacement between the lift pins and the first wafer support so that the lift pins no longer protrude from the first wafer support and the autocalibration wafer rests on the upper surface of the first wafer support.

[0046] In some implementations of the system, the system may include the semiconductor processing tool, and the semiconductor processing tool may include: a wafer handling robot; one or more wafer stations; and a second controller. In such a system, the wafer handling robot and the second controller may be communicatively coupled, and the second controller and the first controller may be further configured to: a) select a first wafer support of the one or more wafer supports of a first wafer station of the one or more wafer stations based at least in part on an indication that an edge ring is supported by the first wafer support; b) cause the auto-calibration wafer to be placed on the edge ring; c) cause each proximity sensor to measure a distance between the first wafer support and the auto-calibration wafer; d) determine one or more height measurements associated with the edge ring based on the one or more distances; e) evaluate the one or more height measurements to determine whether a height associated with the edge ring exceeds a predetermined threshold; and f) provide a notification when the height associated with the edge ring exceeds the predetermined threshold. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Various implementations disclosed herein are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings in which like reference numerals refer to similar elements.

[0048] Figure 1 An example of a correction wafer positioned relative to an edge ring is shown.

[0049] Figure 2 A schematic diagram of an exemplary auto-calibration wafer is shown.

[0050] Figure 3 A wafer support with lift pins is shown.

[0051] Figure 4 Shown is a side view of an auto-calibrated wafer with an array of proximity sensors that can be used to determine edge ring height.

[0052] Figure 5 A side view of another auto-calibration wafer is shown having two sets of proximity sensors, each set of proximity sensors being positioned along a circular path of a different diameter.

[0053] Figure 6 A photograph of an exemplary automated calibration wafer.

[0054] Figure 7 FIG. 4 is a plan view of another exemplary automated calibration wafer.

[0055] Figures 8a to 8i A schematic diagram of a semiconductor processing tool during various stages of operation is shown.

[0056] Figure 9A flow chart is shown of a technique for determining the location of reference points of a structure at a wafer station using an automatic calibration wafer.

[0057] Figure 10 A flow chart showing a technique for determining the relative position of two structures at a wafer station using an automatic calibration wafer.

[0058] Figure 11 A flow chart showing a technique for determining the location of the center point of a wafer support using an automatic wafer alignment technique.

[0059] Figure 12 A flow chart showing a technique for correcting placement of an edge ring on a wafer support.

[0060] Figure 13 A flow chart of a technique for correcting placement of a wafer relative to an edge ring on a wafer support is shown.

[0061] Figure 14 A flow chart showing a technique for verifying repeatability of wafer placement.

[0062] Figure 15 A flow chart showing a technique for evaluating the height of an edge ring.

[0063] Figure 16 A flow chart showing a technique for evaluating lift pin vibration.

[0064] Figure 17 A flow chart showing a technique for assessing base levelness.

[0065] The figures herein are generally not drawn to scale, but various aspects of the figures, such as discussed below, may be drawn to scale. DETAILED DESCRIPTION

[0066] In a typical semiconductor processing system, accurate wafer placement during preparation for various semiconductor processing operations and / or wafer handling operations is achieved via a manual or semi-automatic teaching process that typically involves a technician or other person overseeing: (1) "teaching" a wafer handling robot regarding the desired positions of a semiconductor wafer, an edge ring, and / or a wafer support such as an electrostatic chuck (ESC) relative to each other or relative to an end effector of the wafer handling robot; (2) "teaching" an active wafer centering (AWC) system; and (3) performing wafer placement repeatability verification. Once the wafer handling robot has been taught such positions, for any given wafer, any potential deviations in wafer placement from such positions due to, for example, slight misalignment of the wafer with the end effector during transfer of the wafer to the end effector can be corrected using the AWC. Such manual or semi-automatic teaching processes are time consuming and difficult to implement.

[0067] Such a teaching process typically begins by teaching the wafer handling robot the various positions at which it will pick up or place wafers during operation of a semiconductor processing tool. Generally speaking, while the wafer handling robot may initially be configured to have a general feel for each such position in the equipment, the wafer handling robot will require some degree of customization to adapt its operation to the specific characteristics of the particular semiconductor processing tool in which it is installed, such as varying component or part tolerances. To accomplish this, the wafer handling robot may be placed into a teaching mode in which, for each position to be "taught," the wafer handling robot may be directed to a position corresponding to that position, or to a configuration that places the wafer handling robot in certain "ideal" conditions for that position—for example, a position or configuration in which the wafer handling robot is positioned when the center of the semiconductor wafer is within a known distance (generally a desired minimum or down to zero) from a reference point on the wafer handling robot's end effector, and also within a known distance (generally a desired minimum or down to zero) from the center of the wafer support at that position.

[0068] In typical wafer handling robot training, an "ideal" positioning of the wafer handling robot for each position is achieved, for example, by using one or more fixtures or other structures that interface with features on the wafer handling robot's end effector that can serve as reference points and with another feature fixed relative to the target position. For example, in some wafer handling robot training scenarios, a disk-shaped fixture having a diameter similar to that of a semiconductor wafer can be placed on the wafer handling robot's end effector using, for example, a shaft or pin that passes through the center of the dish and through a reference feature (e.g., a hole) in the end effector. The disk-shaped fixture is centered at a desired location on the end effector where the semiconductor wafer will be centered when the end effector is transported by the wafer handling robot using the end effector. Similar pins may be mounted on features of the wafer support that, when the end effector / dish is urged against such pins, can contact the edge of the dish, thereby guiding the dish and end effector to a specific position, such as a centered position. During such movement of the wafer handling robot, the wafer handling robot can be placed in an unpowered state, allowing an operator to easily manually move the joints / arm sections of the wafer handling robot. Once the wafer handling robot is properly positioned, the controller of the wafer handling robot can obtain measurements of the various joint positions of the wafer handling robot, such as by obtaining measurements of the relative or absolute angular displacements of the various rotational joints of the wafer handling robot, to determine when the relevant motion state of the wafer handling robot is in a desired position. Once the wafer handling robot obtains such position information and the position associated therewith, the position can be considered to have been taught to the wafer handling robot.

[0069] Once a wafer handling robot has been taught various positions to which it will deliver a wafer (or retrieve a wafer), the wafer handling robot can be controlled to enter the kinematic state for the taught position when performing future wafer transfer operations involving that position. If a semiconductor wafer placed onto the end effector of the wafer handling robot is positioned precisely centered at the end effector's reference point, then after the wafer handling robot returns to its kinematic state for the taught position, the semiconductor wafer will similarly be properly centered at the destination position when the end effector of the wafer handling robot places the same semiconductor wafer. However, due to various factors, a semiconductor wafer placed onto the end effector of the wafer handling robot may not be precisely centered at the end effector's reference point. Such misalignment may be relatively minor, e.g., on the order of tens of microns, but such minor misalignments can be detrimental to wafer handling operations. The use of active wafer centering (AWC) systems to correct for such end effector / wafer misalignments has become commonplace in the industry.

[0070] In a typical AWC configuration, an optical AWC sensor and a beam emitter are positioned in a fixed position outside a semiconductor wafer processing chamber so that as a wafer handling robot transports the semiconductor wafer through the processing chamber, the semiconductor wafer moves through two or more of the multiple beams emitted by the beam emitter. The AWC sensor detects when each beam encounters the edge of the semiconductor wafer (as evidenced by the beam being obscured or reformed toward one of the multiple optical sensors). The AWC system can obtain from the wafer handling robot's sensor a defined coordinate position of a reference point on the wafer handling robot's end effector (e.g., a point nominally centered on the semiconductor wafer) each time the edge of the semiconductor wafer triggers one of the AWC optical sensors. (Since the exact placement of the semiconductor wafer on the end effector may be unknown, this defined coordinate position is only an estimate or expected center position.) For a circular semiconductor wafer and at least two AWC beam sensors, the resulting four or more coordinates (as few as three coordinates may be used) are sufficient to determine the position of the semiconductor wafer's center point relative to the semiconductor processing chamber (both the AWC sensor and the wafer handling robot base are fixedly mounted relative to the semiconductor processing chamber). Once such wafer center information is obtained, it can be used as a reference position for future wafer placements, or it can be used as a measurement of the current wafer position that may need correction.

[0071] For example, to train the AWC system, a reference wafer can be manually centered on a desired destination (e.g., a susceptor) in a semiconductor processing chamber. This manual centering can be performed, for example, with the aid of a fixture or jig that guides the reference wafer to be properly centered relative to the susceptor. Once the reference wafer is deemed adequately centered on the susceptor, a wafer handling robot can be controlled to retrieve the reference wafer and remove it from the semiconductor processing chamber. Upon removal from the semiconductor processing chamber, the center of the reference wafer can be measured and determined using the AWC system. This information, along with information from the wafer handling robot indicating the position the reference wafer experienced as it moved from a centered position on the susceptor to a center position determined by the AWC system, can be used to adjust future wafer placements to achieve similar wafer-to-susceptor centering. For example, if a new wafer is placed on the wafer handling robot and passed by the AWC sensor in a manner similar to that of the reference wafer, it may be observed that the center of the new wafer deviates somewhat from the previously determined center position of the reference wafer, e.g., by 0.5 mm in the X direction and 0.25 mm in the Y direction. To correct for such variations, the wafer handling robot can be controlled to counteract such displacements when a new wafer is placed on the susceptor by applying a correction displacement when the new wafer is placed, such as moving the new wafer an additional -0.5 mm in the X direction and an additional -0.25 mm in the Y direction before, after, or during displacement of the new wafer (opposite of the displacements used to retrieve the reference wafer from the susceptor and move the reference wafer toward and past the AWC sensor).

[0072] Similar techniques using an AWC system can also be used to adjust the placement of an edge ring on a wafer support. For example, the AWC system can be used to determine the center point of the edge ring as it passes through the beam of the AWC system, as well as any offset between such an edge ring center point and, for example, a reference position used by the AWC system as the "ideal" center placement position of the wafer or edge ring. In implementations where an AWC system is used in conjunction with edge ring placement, the edge ring has more edge / beam intersections than the semiconductor wafer because the edge ring has an inner and outer edge (thus, each AWC optical sensor detects four edge / beam intersections as the edge ring passes through, rather than the two intersections detected when the semiconductor wafer passes through the beam sensor). In such implementations, data obtained from certain edge / beam intersections (e.g., intersections of the beam with the outer edge of the edge ring) can be ignored, and the center of the edge ring can be determined based on the remaining edge / beam intersections (e.g., intersections of the beam with the inner edge of the edge ring). In some implementations, the reference point used by the AWC system to assess the extent to which the edge ring deviates from the desired placement position on the end effector of the wafer handling robot can be the same reference point that can be used for AWC correction of the semiconductor wafer, i.e., a reference point determined based on measurements obtained from the semiconductor wafer using the AWC system. In other implementations, the reference point used for AWC correction of the edge ring can be obtained based on measurements obtained from the edge ring using the AWC system, for example, the AWC system can be trained with the edge ring to obtain a reference point for future edge ring placements. It should be understood that the placement techniques discussed herein, as well as the AWC training and correction techniques discussed herein, are generally applicable in the context of both semiconductor wafer placement operations and edge ring placement operations.

[0073] AWC systems are widely used in the semiconductor processing industry and generally provide good wafer centering performance. However, the training process is extremely labor-intensive, time-consuming, and prone to user error. While such training can be performed as part of the initial semiconductor processing tool setup, it may need to be repeated periodically during the tool's life, for example, when any modifications are made to the relative positions of the semiconductor processing chamber, AWC sensors, and wafer handling robot, or if wafer processing operations begin to exhibit non-uniformity suggesting that the placement of the semiconductor wafer on the susceptor has become off-center, or generally when any maintenance (including wet cleaning) is performed on the processing chamber. It should be understood that references in various implementations to placing a wafer or edge ring on a wafer support, susceptor, or other device by a wafer handling robot implicitly include indirectly placing the wafer or edge ring on the wafer support. For example, in many semiconductor processing tools, vertically movable lift pins positioned in the wafer support may be used to lift the wafer off the end effector of the wafer handling robot, which then moves away from underneath the wafer without moving the wafer. The lift pins can then be controlled to lower the wafer onto the wafer support. The same process can be repeated in reverse to remove the wafer or other structure from the wafer support. As used herein, the term "wafer support" may refer to any of a variety of structures that can be used to support a semiconductor wafer within a semiconductor processing tool (excluding the end effector of a wafer handling robot). The wafer support may include, for example, a pedestal, an ESC, or other generally circular pedestal-like structure that can be located within a semiconductor processing chamber (or other cavity) and generally contacts the semiconductor wafer in a distributed manner (e.g., via face-to-face contact between the backside of the semiconductor wafer and the upper surface of the wafer support), or a structure that can support the semiconductor wafer via more restricted contact (e.g., an arc-shaped support frame that can contact the semiconductor wafer at multiple points along the outer circumference of the semiconductor wafer). The wafer support may include not only components that directly contact the semiconductor wafer, but also components or portions that extend beyond the outer circumference of the semiconductor wafer, such as an annular portion of a pedestal or ESC that extends beyond the outer circumference of the semiconductor wafer. In some cases, the structure of the wafer support can provide movement, such as vertical movement and / or rotational movement, to the semiconductor wafer supported thereby. In some cases, the wafer support may also include lift pins as discussed above, or other mechanisms that can extend upward relative to the remainder of such a wafer support, thereby lifting the semiconductor wafer off the remainder of the wafer support. In some cases, the wafer support may include various removable components, such as an edge ring.For example, certain wafer supports may be used with a removable edge ring, e.g., one designed to be removed by a wafer handling robot of a semiconductor processing tool and may be docked with one or more non-removable edge rings, e.g., one designed not to be removed by a wafer handling robot of a semiconductor processing tool - such non-removable edge rings may, of course, still be removable by a technician, and for purposes of this specification, such non-removable edge rings are considered part of the wafer support.

[0074] The present disclosure contemplates an automated calibration system, such as one that can be used in conjunction with an AWC system (or similar device) and / or a wafer handling robot to provide, among other things, an adaptive positioning system for automatically teaching the AWC system and / or wafer handling robot of a semiconductor processing tool. Because the chamber in which the teaching occurs can be sealed, as it is during normal semiconductor processing operations, such a system can be used to automatically teach a wafer handling robot under vacuum or atmospheric pressure. Such an automated calibration system can also enable various aspects of component or wafer placement to be evaluated and corrected as necessary to meet processing requirements. An automated calibration system can also be used to guide the placement of an edge ring, which is a nominally annular structure with an inner diameter that is typically slightly larger (or, in some cases, slightly smaller) than the outer diameter of a semiconductor wafer, thereby effectively "extending" the diameter of the semiconductor wafer during processing. Edge rings have the effect of causing an "edge effect," which can degrade the uniformity of process results occurring at the outer edge of the edge ring, rather than on the semiconductor wafer itself.

[0075] At the heart of the auto-calibration system is an auto-calibration wafer, which may also be referred to as an adaptive positioning system (APS) wafer and which collects a wealth of information from various on-board sensors; this enables the auto-calibration wafer to be used as part of an overall auto-teach process. Such an auto-calibration wafer can be used, for example, to perform diagnostic evaluations of components in a semiconductor processing tool and to obtain information that enables adjustments to the operation of the semiconductor processing tool to improve wafer processing performance.

[0076] Generally speaking, an auto-alignment wafer for a particular semiconductor processing tool can have a size and shape similar to the wafers and / or edge rings that the semiconductor processing tool is configured to process, thereby enabling a wafer handling robot of the semiconductor processing tool to transport the auto-alignment wafer in substantially the same manner as it transports semiconductor wafers during processing. Thus, the auto-alignment wafer can be sized to have a maximum height and diameter that are less than the minimum vertical and horizontal clearances of a channel of the semiconductor processing tool through which the wafer handling robot can transport the wafer.

[0077] As described above, the auto-calibration wafer can include a variety of sensors—although the number and type of sensors can vary depending on the specific functionality provided by the auto-calibration wafer. It should be understood that an auto-calibration wafer according to the present disclosure can be used to provide any, some, or all of the sensors / functionality discussed herein.

[0078] In addition to the various sensors that an automated calibration wafer may include, the automated calibration wafer may also include various components for controlling the sensors and obtaining data from the sensors, communicating with other components (e.g., a controller of a semiconductor processing tool), and / or storing and / or manipulating the data collected from the sensors. Thus, such an automated calibration wafer may be linked to a controller of a semiconductor processing tool, introduced into the semiconductor processing tool, and then, via actions of one or both of the controller (or controllers) of the automated calibration wafer and the controller (or controllers) of the semiconductor processing tool, perform various sensing and data collection operations during various stages of a calibration routine or placement routine performed by the semiconductor processing tool. As will be apparent from the examples discussed in greater detail below, a semiconductor processing tool may perform such a calibration routine or placement routine with minimal or no human oversight.

[0079] Generally speaking, an auto-calibration wafer can have a substrate with an overall shape similar to that of a semiconductor wafer, such as being generally circular, though it should be understood that in some cases, the auto-calibration wafer can have a different shape—for example, portions of the substrate not used to support sensors or other components, or not in contact with contact pads on the end effector or lift pins of the pedestal, can be omitted, resulting in openings or notches in the auto-calibration wafer. Furthermore, in some implementations, the auto-calibration wafer can have peninsulas or other protrusions along a nominally circular outer edge, for example, to support sensors at locations beyond the nominal outer diameter of the corresponding semiconductor wafer. The auto-calibration wafer can also include one or more indexing features, such as flat edges, notches, etc., along the outer edge to provide a mechanism for identifying the orientation of the auto-calibration wafer. Reference herein to the "center" of the auto-calibration wafer should be understood to be the point on the auto-calibration wafer that will be located at the same location as the center of a semiconductor wafer or edge ring when the auto-calibration wafer is transported or positioned within a semiconductor processing tool in a manner similar to that typically used to transport or position semiconductor wafers or edge rings. It should be understood that while the center of the auto-calibration wafer may coincide with the geometric center of the substrate and / or the center of mass of the auto-calibration wafer, such alignment is not a requirement.

[0080] In most implementations, the autocalibration wafer may have multiple downward-facing camera sensors, such as charge-coupled device (CCD) sensors or complementary metal-oxide semiconductor (CMOS) sensors, at various locations radially offset from the center of the autocalibration wafer. This offset may be selected so that, when the autocalibration wafer is positioned, for example, above a semiconductor wafer placed on a wafer support of a semiconductor processing tool, each camera sensor has a field of view wide enough to capture a portion of the edge of the semiconductor wafer and a portion of the wafer support. If the semiconductor processing tool typically uses an edge ring during wafer processing operations, the camera sensors may be radially offset from the center of the autocalibration wafer so that the camera sensors' field of view is wide enough to capture a portion of the edge ring when placed on the wafer support. In some autocalibration wafer implementations, the downward-facing camera sensors may be arranged along a circle having a diameter nominally the same size as a semiconductor wafer that the semiconductor processing tool is configured to process, for example, within ±10% or ±20% of the diameter of the semiconductor wafer. In another implementation, the downward-facing camera sensor can be positioned along a circle having a diameter between a semiconductor wafer that the semiconductor processing tool is configured to process and an edge ring that the semiconductor processing tool is configured to use. In some such implementations, the downward-facing camera sensor can be positioned along a diameter that is, for example, an average of, or within ±10% of, a nominal outer diameter of the edge ring and an inner diameter of the edge ring or a nominal outer diameter of the semiconductor wafer. Such positioning, for example, enables the downward-facing camera sensor to simultaneously capture the outer edge of the wafer support (or the edge of a feature on the wafer support, such as an ESC), the outer and / or inner edges of the edge ring, and the outer edge of the semiconductor wafer in its field of view when the autocalibration wafer is positioned approximately centered on the wafer support (and when the semiconductor wafer and / or edge ring are centered on the wafer support).

[0081] Such camera sensor placement enables the automatic calibration wafer to simultaneously obtain images of various fiducials relative to the wafer support and the semiconductor wafer and / or edge ring placed on the wafer support. As used herein, the term fiducial refers to a feature that is assumed to be substantially fixed relative to a particular structure, such as a circular edge of a component that can be used as a fiducial for the center point of that structure (it should be understood that such a fiducial may experience changes in size and / or shape over time, for example, due to erosion or deposition from wafer processing operations; such gradual changes in shape and / or size should not be considered to change the "fixed" nature of such a fiducial in the context of this disclosure). In the examples discussed herein, the fiducials used are features such as the outer edge of a semiconductor wafer, the outer and / or inner edges of an edge ring, the outer edge of a wafer support or the edge of a feature of a wafer support, a surface discontinuity in a wafer support (e.g., the upper surface of a wafer support can have a circular protrusion surrounded by an annular surface that is recessed downwardly from the circular protrusion; the transition between the circular protrusion and the annular surface can be such a surface discontinuity), or any other feature suitable for the techniques discussed herein.

[0082] Images of a given set of fiducials acquired for a semiconductor wafer, edge ring, or other structure can then be analyzed to determine the offset (or offsets) between reference points on two structures associated with those fiducials. For example, if the fiducials used are a circular outer edge of a semiconductor wafer and a circular inner edge of an edge ring surrounding the semiconductor wafer, the relative size of the radial gap between the outer edge of the semiconductor wafer and the inner edge of the edge ring in each image can be determined and used to generate an estimate of the degree to which the center of the semiconductor wafer deviates from the center of the edge ring. Generally, images from at least three image sensors must be used to make such a determination (similar techniques can be implemented using images from only two image sensors, but are generally less accurate because some assumptions must be made regarding the relative sizes of the fiducials used). In some cases, certain fiducials may not always be visible, in which case intermediate fiducials may be required. For example, if a semiconductor wafer is centered on an ESC having an outer diameter smaller than the wafer diameter, and the outer edge of the ESC is used as a reference for such wafer placement, an image may be obtained from the camera sensor of the autocalibration wafer in which both the outer edge of the ESC and the outer edge of the semiconductor wafer are visible—thus, any radial gap between such edges cannot be identified and the degree of centering of the semiconductor wafer on the ESC cannot be directly determined. In such a situation, an intermediate reference, such as the inner edge of an edge ring, can still be used to determine an estimate of the degree of centering of the semiconductor wafer on the ESC. For example, if an edge ring is placed on a wafer support so that it surrounds the ESC, the autocalibration wafer can be used to determine the radial gap between the inner edge of the edge ring and the outer edge of the ESC. Based on the radial gap, the center-to-center offset between the inner edge of the edge ring and the outer edge of the ESC can then be determined, and if necessary, the edge ring can be repositioned until it is acceptably centered on the ESC. After the edge ring is centered, the inner edge of the edge ring can serve as a "proxy" for the outer edge of the ESC. When a semiconductor wafer is subsequently placed on the wafer support and within the edge ring, an image of the radial gap between the outer edge of the semiconductor wafer and the inner edge of the edge ring can be obtained using a camera sensor that automatically aligns the wafer. Because the inner edge of the edge ring is centered on the outer edge of the ESC, the center-to-center offset between the outer edge of the semiconductor wafer and the inner edge of the edge ring can serve as a proxy for the center-to-center offset between the outer edge of the semiconductor wafer and the outer edge of the ESC. In this case, the center-to-center offset between the outer edge of the semiconductor wafer and the inner edge of the edge ring can be further fine-tuned based on the actual measured center-to-center offsets between the outer edge of the semiconductor wafer and the inner edge of the edge ring, and between the inner edge of the edge ring and the outer edge of the ESC.For example, if the edge ring is ultimately positioned so that there is an XY offset (15 μm, 10 μm) between the center of the inner edge of the edge ring and the center of the outer edge of the ESC that is within an acceptable range of centrality and the semiconductor wafer is then placed so that the center of the outer edge of the semiconductor wafer has an XY offset (-5 μm, 12 μm) from the center of the inner edge of the edge ring (in the same coordinate system), the XY offset between the center of the semiconductor wafer and the center of the outer edge of the ESC can be determined by summing these two offset pairs, e.g., (15 μm-5 μm, 10 μm+12 μm)=(10 μm, 22 μm), e.g., a total offset of approximately 24 μm.

[0083] Once the fiducials of the two structures are imaged using the automated calibration wafer and the offset between the two structures is determined, the offset can be compared to an offset threshold, which can represent a maximum offset between the two structures that is acceptable for proper operation of the semiconductor processing tool. If the offset exceeds the offset threshold, the semiconductor processing tool can be prompted to take corrective action.

[0084] While various techniques may be used to determine the center-to-center offset between two structures and are considered to be within the scope of the present disclosure, Figure 1 An exemplary technique for determining such a shift is provided in the context of FIG. Figure 1 An example showing the calibration wafer positioned relative to the edge ring. Figure 1 , an edge ring 162 is shown with a calibration wafer 160 positioned within the inner diameter of the edge ring 162. For clarity, the term calibration wafer or reference wafer as used herein refers to a wafer having dimensions equal to or similar to typical semiconductor wafers processed by a semiconductor processing tool—a calibration wafer intended to replace, for example, a normal wafer that a semiconductor processing tool may process. The calibration or reference wafer may, in some cases, include calibration marks or other features that can be recognized by machine vision algorithms to assist in centering and calibration operations. The term auto-calibration wafer as used herein refers to a "smart" wafer or a wafer that is otherwise equipped with sensors and other electronics to enable such an auto-calibration wafer to acquire data and measure various parameters related to the performance of a semiconductor processing tool.

[0085] Figure 1The calibration wafer 160 and edge ring 162 are not shown to scale in order to more easily illustrate, for example, the misalignment between the edge ring 162 and the center of the calibration wafer 160 (shown as a cross) and the gap between the outer edge of the calibration wafer 160 and the inner edge of the edge ring 162. The calibration wafer 160 can include a set of calibration marks 170, which in this example can be located along the outer circumference of the calibration wafer and can be spaced apart by a known amount (e.g., 120°). A downward-facing first camera sensor of the automatic calibration wafer can be positioned so that the calibration marks 170 and the gap 164 between the edge ring 162 and the calibration wafer 160 both fall within the field of view 128 of the first camera sensor.

[0086] In such a technique, certain assumptions can be made about various factors—for example, the diameter of the calibration wafer can be assumed to be a known quantity, such as 300 mm. Thus, the radius r from the center point of the calibration wafer 160 to the edge of the calibration wafer 160 can be assumed to be w is constant (there may be portions along the edge of the calibration wafer that have a shorter radius, for example, there may be index flats or notches along the edge, but the calibration wafer may have a constant radius in the field of view of the first imaging sensor). As described above, the calibration wafer may also have, for example, calibration marks 170, the positions of which may be very well defined. In this example, the known calibration marks 170 are spaced 120° apart around the circumference of the calibration wafer. The calibration marks may include, for example, features such as radial lines (which are identifiable in the image data and used to establish the direction vector along which the gap 164 is evaluated in each image) and may have a known size, such as 2 mm (as shown) or a circle, which known size may be used to determine the scale of the imaged feature. For example, if a gap 164 is identified in an image as having a size that is 0.23 times the size of the edge of a 2 mm calibration mark square in the same image, then the size of the gap is determined to be 0.23*2 mm = 0.46 mm. In some implementations, such calculations may take into account image distortions due to lens effects (e.g., using a checkerboard pattern), for example, using machine vision or optical image correction techniques, and may correct for such distortions using, for example, calibration data associated with an auto-calibration wafer.

[0087] It should be understood that in some implementations, the calibration marks discussed above may be omitted, and fiducials such as circular edges of various components may be used without determining the center position of such components.

[0088] If it is assumed that the center of the calibration wafer 160 also serves as the origin of the coordinate system in which the center-to-center offset of the edge ring 162 from the calibration wafer 160 is to be determined, the edge ring's center position can be solved by determining the positions of three points along the inner (or outer) edge of the edge ring 162 and then determining the center position of the circle defined by these three points. Once the center position of the edge ring 162 is known in the calibration wafer's coordinate system, extracting the center-to-center offset between the two components is a simple process.

[0089] The gap 164 and the radius r of the calibration wafer 160 may be used w And measure the angle along which each gap 164 is located to determine the positions of three points along the inner edge of the edge ring 162. For example, for gap 164 δ1, gap δ1 can be added to radius r w to obtain the radial distance from the calibration wafer origin / center to the inner edge of edge ring 162. The XY coordinate pair of the point where the gap ends at the inner edge of edge ring 162 can be determined using trigonometric relationships based on the angular position of a radius extending from the center of the calibration wafer to the point where the gap ends at the inner edge of edge ring 162. In this example, the angular position of this radius is 0 degrees. Therefore, the XY coordinates of such a point can be determined according to the following equation: X=(r w +δ x )·sin(θ) Y=(r w +δ x )·cos(θ) where δ x is the associated gap distance, r w is the calibration wafer radius, and θ = the angle between the radius extending from the calibration wafer origin to the gap and the polar axis extending from the origin.

[0090] So, for example, if r w =150mm, δ1=17.338mm, δ2=22.823mm, and δ3=37.69mm, then:

[0091] It should be understood that the dimensions provided above are based on Figure 1 The scale of δ is not necessarily large, and such a value of δ is impractically large for typical semiconductor processing tools and wafer handling robots. In actual operation, the achievable value of δ can generally be less than about 1 mm, such as less than 800 μm.

[0092] Once the three coordinate pairs of XY coordinates for the location along the inner edge of the edge ring are known, the location of the center of the edge ring 162 relative to the origin of the coordinate system (the center of the calibration wafer) can be determined using the following equation: Where (x1, y1), (x2, y2) and (x3, y3) are coordinate pairs, and (x c ,y c ) is the coordinate pair of the center of the edge ring. Therefore, in the current example, (x c ,y c ) = (-8.378 mm, -8.618 mm). After establishing the center offset between the edge ring and the calibration wafer, appropriate action can be taken to reduce the center offset between the two components. For example, in the present case, the edge ring can be retrieved by the wafer handling robot and then moved in a manner that reverses the offset, such as (+8.378 mm, +8.618 mm), to center the edge ring on the calibration wafer. Alternatively, the calibration wafer can be moved by the offset (-8.378 mm, -8.618 mm) to center the calibration wafer on the edge ring. If the automated calibration wafer coordinate system is not aligned with the coordinate system used by the wafer handling robot, the center-to-center offset obtained using the automated calibration wafer can be converted to an equivalent offset in the coordinate system used by the wafer handling robot before using the wafer handling robot to correct, for example, the placement of the edge ring. Once the calibration wafer is acceptably centered, the calibration wafer can then be used to train the AWC of the semiconductor processing tool. Various techniques related to using an auto-alignment wafer to facilitate centering operations in semiconductor processing tools will be discussed in more detail below, but initially a general description of the various features of the auto-alignment wafer will be presented.

[0093] Figure 2 A schematic diagram of an exemplary automated calibration wafer is shown with the dashed / shaded area showing the wafer support, edge ring, and calibration wafer positioned thereunder. Figure 2 2 shows an automatic calibration wafer 200 comprising a substrate 202 on which a plurality of different sensors and other electrical components are mounted. Figure 2, although the calibration wafer 260, annular edge ring 262, and wafer support 252 are not part of the automatic calibration wafer 200; these additional components are shown concentrically with the automatic calibration wafer 200, as during certain stages of normal use, the calibration wafer 260, edge ring 262, wafer support 252, and automatic calibration wafer 200 are all centered relative to each other. In the example shown, the diameter of the automatic calibration wafer 200 is shown to be larger than the wafer support 252, edge ring 262, and calibration wafer 260 - in actual practice, the dimensions of the automatic calibration wafer 200 may be similar to the dimensions of the calibration wafer 260. As previously discussed, the wafer support 252 may include multiple components, such as an ESC that may be slightly smaller than the diameter of the calibration wafer 260, and a support structure that may extend beyond the ESC and support the edge ring 262. For the sake of simplicity, such separate structures are not shown in FIG. Figure 2 middle.

[0094] is displayed as Figure 2 The sensors of the exemplary automatic calibration wafer portion may include, for example, a plurality of first imaging sensors 222, which may be, for example, CCD or CMOS devices. The first imaging sensors 222 may be used in conjunction with an optical or other focusing system, and may be arranged to provide a downward viewing field. Figure 2, the field of view 228 of the first imaging sensor 222 is shown as a rectangular area spanning the outer edge of the calibration wafer 260, the inner and outer edges of the edge ring 262, and the outer edge of the wafer support 252 when the automated calibration wafer 200 is positioned at a predetermined height or range above the calibration wafer 260, the edge ring 262, and the wafer support 252 (e.g., the height at which the end effector of a wafer handling robot normally positions the wafer when transferring the wafer to the wafer support 252). The field of view 228 of each first imaging sensor 222 can be any of a variety of shapes, such as circular or elliptical, and can also extend radially outward to a lesser extent than shown. For example, in some implementations, the field of view 228 of the first imaging sensor 222 may extend only far enough to capture the inner edge of the edge ring 262, but not the outer edge of the edge ring 262. By positioning first image sensor 222 so that it is substantially directly above the outer edge of calibration wafer 260 and the inner edge of edge ring 262 (if used), first image sensor 222 can be positioned to obtain image data that accurately reflects the size of any gaps that may exist between various fiducials in the image (such as the edges of such components). In particular, such image sensor placement can reduce the impact that height mismatches can have on gap size determination, resulting in more accurate gap size estimates. For example, if the image sensor obtains gap image data along a line of sight at a very narrow angle relative to the auto-calibration wafer (such as when the image sensor is mounted near the center of the auto-calibration wafer), any slight variations in the height of either the edge ring or the calibration wafer can be magnified, causing the gap size to fluctuate in an unpredictable manner. Positioning the image sensor near the outer edge of the auto-calibration wafer significantly reduces the impact of such effects.

[0095] In addition to the first image sensor 222, in some cases, the auto-calibration wafer 200 may also include an image sensor, such as a centrally located, downward-facing second image sensor 224. The second image sensor 224 can be configured to obtain an image directly below the center of the auto-calibration wafer. Such an image sensor can be used during a calibration routine where a fiducial to be imaged is located near the location where the center of a semiconductor wafer is typically placed. For example, a wafer support for receiving the wafer may have a cross-shaped fiducial mark located at the center of the wafer support. The cross-shaped fiducial mark can be imaged by the second image sensor to assist in locating the center of the wafer support relative to the auto-calibration wafer. For example, the wafer support may have a fiducial located in the middle. When an end effector of a wafer handling robot positions the auto-calibration wafer 200 over the wafer support, the fiducial can be imaged by the second image sensor 224 to facilitate teaching the wafer handling robot the position of the wafer support. Such a sensor can also be used to center the auto-calibration wafer on the end effector of the wafer handling robot. For example, the end effector of a wafer handling robot may include a fiducial positioned approximately at a location visible to the second camera sensor 224 when the auto-calibration wafer 200 is approximately centered above the end effector. When the wafer handling robot is actuated to retrieve the auto-calibration wafer, the end effector may be moved beneath the auto-calibration wafer so that the fiducial falls within the field of view of the second camera sensor 224. The second camera sensor 224 may then be caused to acquire an image of the fiducial, which may be analyzed to determine how far off-center the fiducial is relative to the center of the auto-calibration wafer 200. The wafer handling robot may then be caused to reposition the end effector to reduce the off-centering of the fiducial relative to the auto-calibration wafer 200 to acceptable limits. It should be understood that other implementations of the auto-calibration wafer 200 discussed herein may utilize sensors other than camera sensors to obtain radial clearance data and / or centering data. For example, an ultrasonic sensor may be used to obtain a contour map displaying the three-dimensional fiducial. The distances and clearances described above may be determined from the contour map in a manner similar to the manner in which such distances and clearances are determined from the camera data. It should be understood that any sensor that can be used to assess the gap between the fiducials described above and / or automatically calibrate the centering of the wafer 200 relative to the wafer support may be used in place of the camera sensor discussed above.

[0096] Some implementations of the auto-calibration wafer 200 may also include various non-camera sensors, such as one or more vibration sensors 230 , one or more orientation / tilt sensors 232 , and / or one or more proximity sensors 234 .

[0097] Vibration sensors 230 can be used to detect vibrations experienced by wafers during various operations, such as wafer handling robot operation or lift pin retraction or extension, for automated wafer alignment. For example, in some wafer stations, the wafer support can be a pedestal or similar structure and can include multiple (e.g., three) lift pins, which are thin pins that can move vertically relative to an electrostatic chuck (ESC) or other wafer support structure. These lift pins are typically arranged in an equilateral triangle shape within a circular boundary defined by a semiconductor wafer centered on the wafer support. Thus, when the lift pins are in an extended position relative to the surface of the wafer support, they can support any wafer present at the wafer station. When the lift pins are retracted into the wafer support, the supported wafer is brought into contact with the upper surface of the wafer support. During such lift pin retraction, the wafer can experience small vibrations, such as those caused by wear and tear on the equipment. One or more vibration sensors can be used to assess the nature of such vibrations and provide an indication of the health of the lift pin mechanism. The shock sensor may include, for example, an accelerometer, a piezoelectric shock sensor, an optical distance measurement sensor, or an optical microphone (such sensors may detect shock, for example, by measuring the displacement of the substrate of an autocalibration wafer relative to the wafer support, thereby providing insight into the extent of the shock experienced), as well as other types of sensors.

[0098] One or more orientation / tilt sensors 232 can be used to assess whether the wafer support or other components are supporting the autocalibration wafer (and therefore other wafers) in a level manner. For example, if one of the lift pins of the wafer support becomes shorter or longer than the other lift pins, the wafer supported by the lift pins may exhibit a slight degree of tilt. Such tilting can cause one side of the wafer to contact the wafer support before the other side of the wafer, which can slightly alter the way the wafer rests on the wafer support and, in some cases, cause some slippage between the lift pins and the wafer during placement on the wafer support (which can damage the wafer and / or cause particle contamination). Furthermore, such sensors can be used to assess the levelness of the wafer support itself after the autocalibration wafer has been placed on the wafer support. Thus, an autocalibration wafer with orientation sensors can be used to assess the levelness of both the wafer support and other equipment. The orientation sensors can include, for example, accelerometers and tilt sensors or inclinometers.

[0099] Figure 3 Wafer support 352 is shown with extended lift pins 372, which are shown supporting automated alignment wafer 300. Lift pins 372 can be retracted downward (and / or wafer support 352 moved upward) to allow automated alignment wafer 300 to be placed on wafer support 352.

[0100] One or more proximity sensors 234 may be used to assess the height of various structures at the wafer station. For example, it may be desirable to assess the height of the edge ring or the portion of the edge ring that is above the wafer support (e.g., above the ESC of the wafer support, around its circumference) (non-uniform circumferential height of the edge ring may develop or increase process non-uniformity). If proximity sensors are provided in the autocalibration wafer and positioned to allow direct or indirect distance measurement between the autocalibration wafer and the edge ring (or other structure), the resulting data may be used to determine the degree of uniformity of the height of the edge ring or portion thereof along its circumference.

[0101] In other implementations of an auto-calibration wafer, the auto-calibration wafer can be placed onto an edge ring so that it can be supported by the edge ring. In some such implementations, the auto-calibration wafer can have one or more portions extending beyond the inner diameter of the edge ring and resting on the uppermost surface of the edge ring (although the semiconductor wafer intended for use with the edge ring will typically be completely contained within the inner diameter of the edge ring). In other such implementations, however, the auto-calibration wafer can be sized to have a diameter similar to that of the semiconductor wafer configured for use with the edge ring. In some such implementations, the edge ring can have a stepped inner diameter; for example, the upper surface of the edge ring can have a diameter slightly larger than the diameter of the semiconductor wafer intended for use with the edge ring, while the lower surface of the edge ring can have a diameter slightly smaller than those semiconductor wafers. The resulting geometry is a concave annular surface in the edge ring that can be used to support the semiconductor wafer during processing. Consequently, an auto-calibration wafer supported by such an edge ring has a small gap between it and the wafer support supporting the auto-calibration wafer. Proximity sensors can be used to determine the size of this gap at various locations along the circumference of the auto-calibration wafer / edge ring. The resulting measurements may be analyzed to determine levelness or thickness variations in the portion of the edge ring between the wafer support and the autocalibration wafer.

[0102] Figure 4 Shown is a side view of an auto-calibrated wafer with an array of proximity sensors that can be used to determine edge ring height. Figure 4In FIG. 4 , edge ring 462 is positioned on wafer support 452. Auto-calibration wafer 400 is positioned so as to rest on the circumferential shelf of edge ring 462, thus slightly overhanging wafer support 452. As can be seen, the edge ring has a certain height within the non-uniform shelf region—the right side of the shelf is higher than the left side, causing auto-calibration wafer 400 to have a tilt angle relative to wafer support 452 and edge ring 462. The proximity sensors 434 in auto-calibration wafer 400 can be configured to measure the distance between each proximity sensor 434 and the nearest facing surface, such as the upper surface of wafer support 452. In this case, the left proximity sensor 434 has measured a distance Δ1, while the right proximity sensor 434 has measured a distance Δ2. These distances can be evaluated against one or more conditions to determine whether the edge ring height exceeds an allowable value. For example, in some implementations, if |Δ1-Δ2|>x or max(Δ1,Δ2)>y, the edge ring may be considered to have an out-of-limit edge ring height and corrective action may be taken, such as requiring the installation of a new edge ring.

[0103] In some such implementations, two sets of proximity sensors are provided on the auto-calibration wafer—one set positioned to obtain distance measurements between the auto-calibration wafer and the edge ring, and the other set positioned to obtain distance measurements between the auto-calibration wafer and non-edge ring structures, such as a calibration wafer placed in the middle of the edge ring or on a surface of a wafer support. In such implementations, the auto-calibration wafer can be positioned directly above the edge ring, for example, supported by lift pins or a wafer handling robot, and proximity sensors can be used to obtain distance measurements between the auto-calibration wafer and the edge ring, and between the auto-calibration wafer and other structures. In such implementations, the auto-calibration wafer can be sized to be larger than a typical semiconductor wafer used in semiconductor processing tools (or to have portions protruding beyond the diameter of the semiconductor wafer) so that it radially overlaps the edge ring with sufficient margin to allow certain proximity sensors mounted thereon to radially overlap the edge ring and determine the distance between the upward-facing surface of the edge ring and those proximity sensors. Other proximity sensors may be located on the auto-calibration wafer so that it radially overlaps the center opening of the edge ring, such that these proximity sensors can obtain distance measurements between the auto-calibration wafer and, for example, a calibration wafer placed in the center of the edge ring, or an exposed surface of a wafer support if a similar wafer does not exist.

[0104] By simultaneously obtaining measurements from both sets of proximity sensors and subtracting, for example, the autocalibration wafer / edge ring distance from the corresponding autocalibration wafer / wafer support or autocalibration wafer / calibration wafer distance, an estimate of the edge ring height at each proximity sensor location can be determined.

[0105] Figure 5 A side view of another auto-calibration wafer is shown with two sets of proximity sensors, each set of proximity sensors being arranged along a circular path of a different diameter. Figure 5 , the end effector 558 of the wafer handling robot supports the autocalibration wafer 500 above the wafer support 552 and the edge ring 562. The outermost proximity sensor 534 is located at a radial position that radially overlaps the edge ring 562 when the autocalibration wafer 500 is nominally centered above the edge ring 562, and the innermost proximity sensor 534 is located at a radial position that radially overlaps the inner portion of the edge ring 562. The proximity sensors 534 can be controlled to simultaneously determine the distance between the proximity sensor 534 and the surface directly below the proximity sensor, for example, the distance Δ 1a , Δ 2a , Δ 1b , and Δ 2b These distances can be evaluated to determine if the edge ring height varies beyond an acceptable range. For example, if |Δ 2a -Δ 2b |>x or max(Δ 2a ,Δ 2b )>y, or if |(Δ 1a -Δ 1b )-(Δ 2a -Δ 2b )|>x or max(((Δ 1a -Δ 1b ),(Δ 2a -Δ 2b ))>y, the height of the edge ring can be considered to have exceeded an allowable threshold. In some such implementations, this measurement can be used to create a closed-loop system in which edge ring lift pins can be actuated to adjust the edge ring height above the wafer support (e.g., above the ESC) to maintain the edge ring height above the wafer support (e.g., above the ESC) at an optimal value. Similar techniques can be used to initially calibrate the edge ring lift pins, for example, to determine the respective heights at which each edge ring lift pin must be located in order to maintain the edge ring level relative to the wafer support. For clarity, it should be noted that a wafer support can have multiple sets of lift pins—e.g., one set can include lift pins located within the region of the wafer support where the semiconductor wafer is to be placed, while another set can include lift pins located outside of that region but within the region of the wafer support occupied by the edge ring. Each set of lift pins can be separately actuated to raise or lower the semiconductor wafer, or to raise and lower the edge ring.

[0106] Various types of proximity sensors may be used, including, for example, capacitive distance sensors, inductive distance sensors, optical distance sensors, etc. In some cases, the auto-calibration chip may include one or more other types of sensors, such as temperature sensors, pressure sensors, humidity sensors, light sensors, and the like.

[0107] The various sensors included in the auto-calibration wafer can be communicatively coupled to a first controller 208, which can include one or more first processors 210 and one or more first memories 212. The first controller 208 can also be electrically coupled to a power source 214, such as a battery, a capacitive battery (capattery), or other power source. In some implementations, the power source 214 can be operatively coupled to a charging feature using electrical contact pins positioned to align with a charging feature at a docking station for storing the auto-calibration wafer 200 when the auto-calibration wafer 200 is placed in the docking station. Figure 2 The illustrated implementation shows a wireless charging feature 216, which can be, for example, an inductive charging coil, such as a Qi compatible inductive charging coil or other suitable wireless charging interface. In such a case, the docking station for storing the automatic calibration wafer 200 can have a similar wireless charging interface configured to charge the automatic calibration wafer 200 when it is placed therein.

[0108] The first controller 208 can also be communicatively coupled to a first wireless communication interface, such as WiFi, Bluetooth, or other wireless communication interface, so that instructions and / or data can be sent from and / or to the first controller 208, and thus from and / or to the auto-calibration chip 200. For example, a semiconductor processing tool interfaced with the auto-calibration chip 200 can include a second controller having one or more second processors and one or more second memories. The second controller can be communicatively coupled to a second wireless communication interface, which can then be used to interface with the first wireless communication interface of the auto-calibration chip. Thus, the auto-calibration chip 200 can be capable of wirelessly communicating with the semiconductor processing tool, enabling the transfer of information, instructions, and other data between the auto-calibration chip 200 and the semiconductor processing tool.

[0109] Figure 6 is a photograph of an exemplary auto-calibration wafer. Auto-calibration wafer 600 includes a substrate 602 having printed circuit lines that provide electrical connections between various components, including a power source 614, which in this example is a rechargeable battery, a processor 610, a memory device 612, and a wireless charging feature 616 that can be used to inductively transfer energy to the rechargeable battery during wireless charging. Figure 6Also visible are three first imaging sensors 622 mounted at equidistant positions along the circumference of substrate 602 and a centrally mounted second imaging sensor 224. Located approximately adjacent each first imaging sensor 622 is a corresponding proximity sensor 634, which in this example is a capacitive proximity sensor.

[0110] Figure 7 FIG2 is a line drawing of a plan view of another exemplary auto-calibration wafer 700. In this example, auto-calibration wafer 700 has a generally circular substrate 702 with three ears dispersed along its outer circumference, each ear housing a corresponding first imaging sensor 722. In this example, substrate 702 has approximately the same diameter as a typical semiconductor wafer; the ears extend beyond this diameter and enable first imaging sensor 722 to be positioned such that its photosensitive area is centered, for example, above the outer edge of a semiconductor wafer positioned below and centered on auto-calibration wafer 700. In this example, auto-calibration wafer 700 also includes two power sources 714, such as rechargeable batteries that can be charged using a wireless charging feature 716 (e.g., an inductive charging coil). The battery can power various electrical components of the auto-calibration wafer 700, such as the processor 710, memory 712, wireless communication interface 718, first camera sensor 722, second camera sensor 724, proximity sensor 734, and accelerometer 736 (which can be used as an orientation or tilt sensor).

[0111] Figures 8a to 8i A schematic diagram of a semiconductor processing tool is shown during various stages of using an automated calibration wafer. Figure 8a A portion of a semiconductor processing tool is shown in . The portion of the semiconductor processing tool shown includes two wafer stations 844a and 844b, but the tool may also include other wafer stations. Each wafer station corresponds to a location where one or more wafers may be placed during various operations performed by the semiconductor processing tool. The wafer station may be, for example, but not limited to, present in one or more processing chambers of the tool, in a buffer for storing wafers before or after processing, in an air lock or load lock that enables wafers to be transferred between environments of different pressures, in a load port, in a front opening unified unit (FOUP) that can be docked to a load interface, etc. In . Figure 8aIn the embodiment shown, wafer station 844a is provided by semiconductor processing chamber 850; conversely, wafer station 844b is provided by a docking station 868 specifically designed for storing auto-calibration wafer 800 (although some implementations may not include such a dedicated docking station). Docking station 868 may include a feature (not shown) for charging auto-calibration wafer 800 or may be otherwise configured to interface with various aspects of auto-calibration wafer 800. In some implementations, the docking station may be located in (or attached to) a vacuum transfer module (VTM) so that it can be accessed by a wafer handling robot in the vacuum transfer module, which can then be trained with the auto-calibration wafer. In other implementations, the docking station may be located in an equipment front end module (EFEM) or other atmospheric or near-atmospheric location, in which case the auto-calibration wafer can be retrieved by a wafer handling robot located in the EFFM and then transferred to another wafer handling robot located in the VTM.

[0112] Each wafer station 844 may have an associated wafer support 852, such as wafer support 852a / pedestal 854 of wafer station 844a (a wafer support is not shown for wafer station 844b, but it may have a wafer support capable of receiving the autocalibration wafer 800 when the autocalibration wafer 800 is placed in the wafer station). In some cases, the wafer station may be associated with an active wafer centering (AWC) system 866, which can obtain measurements of the center position of the wafer as the wafer is introduced into or removed from the associated wafer station 844. In this example, the AWC system 866 is associated with wafer station 844a and includes two vertically oriented beam sensors (represented by dots within the AWC system 866) that can detect when the edge of the wafer crosses either beam. As previously discussed, the AWC system 866 can be used to determine the center position of a wafer supported by the end effector 858 of the tool's wafer handling robot 856 relative to a specific known reference frame, thereby determining whether any positioning corrections are required before placing the wafer at a desired location.

[0113] like Figure 8a As shown in FIG, in preparation for placing the edge ring 862 onto the wafer support 852a / pedestal 854, the wafer handling robot 856 supports the edge ring 862 on the end effector 858. During this time, the automated calibration wafer 800 is temporarily stored in the wafer station 844b / docking station 868.

[0114] exist Figure 8b , the wafer handling robot 856 has been actuated to place the edge ring 862 into a position nominally centered on the wafer support 852a / pedestal 854, and the wafer handling robot 856 has been further actuated to retrieve the automated calibration wafer 800 from the wafer station 844b / docking station 868.

[0115] exist Figure 8c , the wafer handling robot 856 has retrieved the auto-alignment wafer 800 from the wafer station 844b / docking station 868 and is ready to position the auto-alignment wafer over the wafer support 852a / pedestal 854 and edge ring 862.

[0116] exist Figure 8d In the embodiment shown in FIG. 8 , wafer handling robot 856 has extended end effector 858 to position auto-calibration wafer 800 over wafer support 852a / pedestal 854 and edge ring 862. Auto-calibration wafer 800 is thus positioned such that the field of view of a downward-facing first imaging sensor of auto-calibration wafer 800 (represented by three rectangular dots spaced along the outer edge of auto-calibration wafer 800) encompasses edge ring 862 and one or more features of wafer support 852a / pedestal 854. Second controller 842 can then cause the first controller of auto-calibration wafer 800 to obtain image data from the first imaging sensor, such as the gap between the inner edge of edge ring 862 and the outer edge of a fiducial of wafer support 852a / pedestal 854, such as the ESC of wafer support 852a. As previously discussed, these gaps can be used to determine the longitudinal offset of edge ring 862 relative to the center of wafer support 852a / pedestal 854. If this offset exceeds an allowable threshold, the edge ring can be repositioned to reduce the offset. In this example, the edge ring has been placed in an acceptable manner and as Figure 8e The wafer handling robot 856 shown in FIG. 844 can remove the automatic calibration wafer 800 from the wafer station 844 a and return the automatic calibration wafer 800 to the wafer station 844 b .

[0117] exist Figure 8f In FIG. 8 , the wafer handling robot has been activated to retrieve the calibration wafer 860, which may also be stored in the docking station 868 and, for example, below or above the automated calibration wafer 800, or may be obtained from a completely different location such as a load lock or air lock. The calibration wafer 860 may then be placed into the wafer station 844a / semiconductor processing station 850 and transferred to the wafer support 852a / pedestal 854 so that it can be positioned as shown in FIG. Figure 8g 862 is shown nominally centered relative to the center of edge ring 862.

[0118] exist Figure 8h In the embodiment, the wafer handling robot has been activated to retrieve the automatic calibration wafer 800 from the wafer station 844b / docking station 868; Figure 8i , the wafer handling robot has been further actuated to position the automated calibration wafer 800 over the wafer support 852a / pedestal 854, calibration wafer 860, and edge ring 862. Figure 8d, the automatic calibration wafer 800 can then be controlled to obtain image data of a gap, such as gap 864, between the edge ring 862 and the calibration wafer 860, thereby determining any offset between the center 860 of the calibration wafer and the center of the edge ring 862.

[0119] The following will refer to Figures 9 to 14 Various techniques that can be implemented with an auto-calibration wafer, such as those discussed above, are explored in more detail.

[0120] Figure 9 A flow chart illustrating a technique for determining the location of a reference point of a structure at a wafer station using an autocalibration wafer. Such techniques can be used, for example, to teach a wafer handling robot various positions where a wafer may be placed (or retrieved). In block 902, the autocalibration wafer can be retrieved using the end effector of the wafer handling robot; in block 904, the autocalibration wafer can be centered on the end effector of the wafer handling robot (blocks 902 and 904 can be performed simultaneously in some implementations). For example, the autocalibration wafer can be placed on the end effector such that the center of the autocalibration wafer (or some other known reference point on the autocalibration wafer) is centered on a known reference point on the end effector, thereby establishing a spatial relationship between the two reference points and allowing measurements obtained using the autocalibration wafer to be mapped or converted into the coordinate system used by the wafer handling robot.

[0121] Such placement of the autocalibration wafer onto the end effector can be accomplished via any suitable mechanism, including through the use of physical indexing features or other contact-based means to ensure that the autocalibration wafer is properly positioned on the end effector. However, in some cases, imaging features of the autocalibration wafer itself can alternatively be used to ensure that the autocalibration wafer is centered on the end effector. For example, before the autocalibration wafer is loaded onto the end effector, the autocalibration wafer can be positioned over the end effector and one or more of the autocalibration wafer's imaging sensors can be activated to obtain an image of the end effector or a region thereof. The portion of the end effector imaged can include, for example, a fiducial defining a reference point on the end effector, such as a location on the end effector that coincides with the XY center point of a theoretical semiconductor wafer perfectly positioned on the end effector. This image data can then be analyzed to determine the extent to which a reference point of the autocalibration wafer, such as the center of the autocalibration wafer, is offset from a reference point / fiducial of the end effector. The wafer handling robot may then be actuated to move the end effector in a manner that reduces or eliminates this offset before the auto-calibration wafer is placed on the end effector, thereby centering the auto-calibration wafer on the end effector.

[0122] In block 906, a wafer station can be selected for calibration. Calibration involves, for example, determining a reference point on a wafer station structure, such as a wafer support, where the center of a wafer transferred to the wafer station is expected to be located. In block 908, the wafer handling robot can be actuated to position the end effector and the auto-calibration wafer over the selected wafer station, such that the auto-calibration wafer is approximately centered over, for example, a reference point on the wafer support of the selected wafer station. This initial positioning can be based on, for example, an estimate of the wafer support reference point location, which can be based on the designed positions of various components in the system, and can generally achieve placement accuracy within one or a few millimeters in most cases.

[0123] In block 910, an autocalibration wafer may be caused to acquire image data of one or more fiducials located on a target structure at a selected wafer station. The one or more fiducials may be associated with a reference point of the wafer station, such as the location where the center of a wafer transported to the wafer station is expected to be located. For example, the outer edge of an ESC of a wafer support at the selected wafer station may be used as a fiducial. Such a fiducial may not directly indicate the reference point of the wafer station, but may clearly define the reference point. For example, a circular or curved edge of the ESC may define a center point used as a reference point. In another example, the wafer support may include some type of fiducial, such as an etched "+" or other marking that directly identifies a reference point. For example, the center of the wafer support may be the reference point, and the intersection of two lines within the "+" may represent the reference point.

[0124] In block 912, a position of a reference point of a selected wafer station structure (e.g., a wafer support) associated with the auto-calibration wafer may be determined based on the fiducial image data. For example, the image data may indicate that the reference point of the structure is XY offset from a reference point of the auto-calibration wafer, such as the center of the auto-calibration wafer (0.3 mm, 0.5 mm) in the auto-calibration wafer's coordinate system.

[0125] The position of the reference point of the structure may then be determined relative to the wafer handling robot coordinate system in block 914. For example, the XY offset determined in block 912 relative to the coordinate system of the auto-calibration wafer may undergo a coordinate system transformation to convert it to equivalent coordinates in the coordinate system of the wafer handling robot, e.g., to account for possible angular misalignment between the coordinate systems of the auto-calibration wafer and the wafer handling robot.

[0126] If you use Figure 9In some cases, it may be desirable to calibrate the auto-calibration wafer before performing a calibration technique. For example, it may be desirable to establish the position of an imaging sensor, or the sensor used to obtain image data, relative to a reference point, such as the center point of the auto-calibration wafer, in order to properly process position information determined from such sensors. Each imaging sensor can, for example, be considered to provide XY position data in a coordinate system (based on a rectangular or linear array of pixels that each such sensor may have) that is specific to each imaging sensor and offset from the auto-calibration wafer's reference point by a specific XY distance and / or rotation angle. Calibrating the auto-calibration wafer can determine such XY and angular offsets for each imaging sensor's coordinate system. Any subsequent position determined from the imaging sensor data can be appropriately transformed to be accurately located relative to the coordinate system of the auto-calibration wafer's reference point.

[0127] In one example of such calibration, an autocalibration wafer can be placed in a fixture that has index pins or other alignment features that contact the outer edge of the autocalibration wafer and physically constrain the autocalibration wafer to be centered over a fiducial that is part of the fixture and known to be centered relative to the constrained outer edge of the autocalibration wafer. Once the autocalibration wafer is mounted in the fixture and centered over the fiducial, a centered camera sensor can be made to acquire an image of the fiducial and then, for example, determine which pixel or pixels coincide with the center point indicated by the fiducial, thereby providing information that can subsequently be used to convert any positional data obtained from the image of the centered camera sensor into a coordinate system relative to the reference point. Similar fiducials can be provided in the fixture at locations that coincide with the fields of view of other camera sensors, allowing calibration of all camera sensors before use.

[0128] Figure 10 A flow chart is shown of a technique for determining the relative positioning of two structures at a wafer station using an automatic calibration wafer. Figure 10 The technique begins at block 1002, where an automated calibration wafer is retrieved from a docking station or other storage area for automated calibration wafers using a wafer handling robot of a semiconductor processing tool. At block 1004, a wafer station of the semiconductor processing tool to be calibrated may be selected. Assuming that the structure whose relative positioning is to be determined is already located in the selected wafer station, the selected wafer station may, for example, have an edge ring (first structure) placed on a wafer support (second structure).

[0129] In block 1006, a wafer handling robot may be actuated to position the autocalibration wafer over the wafer support of the selected wafer station. The wafer handling robot may, for example, be positioned so that the autocalibration wafer is nominally centered over the wafer support / edge ring of the selected wafer station, thereby positioning a first imaging sensor positioned along the outer circumference of the autocalibration wafer over the edge ring and wafer support in a manner that enables the first imaging sensor to acquire an image of two fiducial structures (e.g., the inner edge of the edge ring and the edge of a feature of the wafer support, such as the outer edge of the ESC of the wafer support). In block 1008, such images of the autocalibration wafer may be acquired. In block 1010, the images may be analyzed to determine, for example, the size of gaps between the fiducials in each image. For example, an edge-finding algorithm may be used to identify the relative distances of the inner edge of the edge ring and the edge of the wafer support in each image, as well as the gap between each pair of determined edges. The gap between each pair of determined edges may be estimated based on an assumed vertical distance between the first imaging sensor and the imaged structure; such estimates may be somewhat inaccurate but will generally be of similar proportions in each image. In the case where the fiducials for each structure are arranged along a common reference circle, for example, the arcuate outer edge or arcuate inner edge (or a single circular edge thereof) of the edge ring, both radially co-axially, can be used as the fiducial for the edge ring, and the arcuate edges (or a single circular edge thereof) of the wafer support, both radially co-axially, can be used as the fiducial for the wafer support. It will be understood that other fiducials with similar effects can be used, and the techniques discussed herein are generally applicable to any suitable fiducial and to algorithms for determining relative offsets between such structures based on a selected fiducial.

[0130] In block 1010, the offset between reference points (e.g., the centers of the two structures) is determined based on the size of the gap between the imaged fiducials of the two structures. This offset can be compared to an offset threshold to determine whether the decentering of the two structures falls within acceptable limits. This offset threshold can be established based on the non-uniformity requirements of a particular semiconductor processing technology. If the decentering measurement does not fall within acceptable limits, appropriate action can be taken, such as repositioning one of the structures based on the measured center offset and repeating the process until the decentering measurement falls within acceptable limits.

[0131] Figure 11A flow chart illustrating a technique for determining the position of a center point of a wafer support using an auto-calibration wafer is shown. In block 1102, the auto-calibration wafer can be retrieved using an end effector of a wafer handling robot. In block 1104, the position of the auto-calibration wafer relative to the end effector of the wafer handling robot can be determined. In some cases, blocks 1102 and 1104 can be performed sequentially. For example, the wafer handling robot can be controlled such that the end effector picks up the auto-calibration wafer (as discussed above) so that the auto-calibration wafer is centered on a reference point of the end effector, thereby determining the position of the auto-calibration wafer relative to the end effector.

[0132] In block 1106, a wafer station may be selected to determine the center point of its wafer support. In block 1108, the wafer handling robot may be actuated to position the autocalibration wafer over the wafer support of the selected wafer station, such as at a default center position associated with the wafer station.

[0133] In block 1110, one or more imaging sensors of the autocalibration wafer may be used to obtain image data of one or more fiducials, such as a wafer support. Such a fiducial may, for example, be an etched pattern located at the center of the wafer support. Alternatively, the fiducial may be a circular edge of a portion of the wafer support, such as the circular edge of an ESC that defines the center point of the wafer support and is part of the wafer support. In the former case, the fiducial may be imaged using an imaging sensor located near or at the center of the autocalibration wafer. In the latter case, the fiducial may be imaged using an imaging sensor located near the outer edge of the autocalibration wafer.

[0134] In block 1112, the image data can be analyzed to determine the offset between a reference point, such as the center point of the autocalibration wafer, and a reference point of the selected wafer station defined by the fiducial. In block 1114, the offset determined in block 1112 can be converted into the coordinate system of the wafer handling robot. If necessary, the "default" position corresponding to the center point of the selected wafer support can be updated to account for the offset determined in block 1114 (or the default position can be left as is and then adjusted based on the offset for each subsequent wafer placement at the wafer station). In some implementations, the wafer handling robot can then be actuated to move the autocalibration wafer so that the center of the autocalibration wafer is positioned at the updated center position of the wafer support (similar to block 1108). In such implementations, blocks 1110 through 1114 can be repeated, if necessary, to verify that the updated center position is properly centered. If it is found that the updated default position (or a modification of the default position) still produces a center-to-center offset that is not within acceptable limits, the process can be repeated one or more times.

[0135] Although the above discussion focuses on using an autocalibration wafer to determine the absolute position of the center of a wafer support or other structure relative to the coordinate system used by a wafer handling robot, as will be discussed below, an autocalibration wafer can also be used to determine the relative positioning between two components.

[0136] Figure 12 A flow chart illustrating a technique for calibrating the placement of an edge ring on a wafer support is shown. In block 1202, a wafer handling robot of a semiconductor processing tool may be actuated to cause the wafer handling robot to retrieve an edge ring from a wafer station. In block 1204, the wafer handling robot may also be actuated to cause the edge ring to be placed on a wafer support, such as a pedestal, of a selected wafer station of the semiconductor processing tool. Blocks 1202 and 1204 may be optional, as the edge ring may have been manually placed or mounted on the wafer support, or may have been placed or mounted on the wafer support, for example, during another phase of operation.

[0137] In block 1206, the wafer handling robot may be controlled to retrieve the auto-calibration wafer from the docking station or other location. In block 1208, the wafer handling robot may also be actuated to position the auto-calibration wafer over the wafer support (and the edge ring positioned thereon) of the selected wafer station.

[0138] Once the autocalibration wafer is positioned over the wafer support of the selected wafer station, the autocalibration wafer may be caused to obtain image data of a gap between an inner edge of the edge ring and an outer edge of a feature of the wafer support (or other reference), such as an outer edge of an ESC that is part of the wafer support, using an edge camera, such as a first imaging sensor, of the autocalibration wafer in block 1210.

[0139] In block 1212, an estimated offset between the center of the edge ring and the center of the wafer support can be determined based on the relative size of the gap between the fiducials in the image. In block 1214, a determination can be made as to whether the estimated edge ring / wafer support offset exceeds a predetermined offset threshold. If the estimated edge ring / wafer support deviation does exceed the predetermined offset threshold, the technique can proceed to block 1216, where a wafer handling robot can be actuated to remove the automated calibration wafer from its position above the wafer support and return the automated calibration wafer to, for example, a docking station (or some other temporary holding location). Next, in block 1218, a wafer handling robot can be actuated to retrieve the edge ring from the wafer support of the selected wafer station. For example, lift pins can be used to lift the edge ring off the wafer support so that an end effector of the wafer handling robot can be inserted under the edge ring, and the edge ring can then be lowered onto the end effector by retracting the lift pins into the wafer support.

[0140] After the edge ring has been retrieved from the wafer support using the wafer handling robot in block 1218, the wafer handling robot can also be controlled to reposition the edge ring onto the wafer support so that the center of the edge ring is located at a new position that accounts for the edge ring offset, thereby more accurately centering the edge ring and wafer support relative to each other. After block 1218, the technique can return to block 1206 and a further assessment of the edge ring / wafer support center offset can be obtained. If necessary, this portion of the technique can be repeated a threshold number of times, or until the estimated center offset between the edge ring and wafer support falls within a predetermined offset threshold. If the estimated edge ring / wafer support offset is found to fall within a predetermined deviation threshold in block 1214, the technique can proceed to block 1222, where the edge ring placement calibration can be considered complete.

[0141] It will be appreciated that using an autocalibration wafer to assess the relative offset between two structures can be accomplished without requiring or knowing the precise positioning of the autocalibration wafer relative to the end effector of a wafer handling robot. In particular, the techniques discussed herein can be used when the autocalibration wafer is sufficiently centered on the end effector such that the field of view of the autocalibration wafer's first camera sensor is capable of imaging the various gaps between the two structures' fiducials. It will also be appreciated that similar techniques can be used to center other components relative to a wafer support, such as centering the calibration wafer relative to the wafer support.

[0142] For an edge ring, once the edge ring is properly placed on a wafer support, it can generally remain in that position for a number of wafer handling operations. However, for calibration wafer centering, the calibration wafer merely replaces or represents the wafer to be placed in future operations. Therefore, once the calibration wafer has been centered on a desired structure, such as a wafer support or edge ring, using the automated calibration wafer techniques discussed herein, the calibration wafer can be removed from the centered position using a wafer handling robot and used to teach an active wafer centering (AWC) system. The AWC system can then learn the desired center point of the calibration wafer relative to the end effector for a particular wafer handling robot's motion path. The active wafer centering system can then be used to evaluate future placements of wafers on the end effector of the wafer handling robot to determine the offset between the center point of such wafers and the learned center point. The wafer handling robot can then be actuated to place the wafer on the wafer support in a manner that accounts for this determined offset. It should be understood that while the edge ring may generally remain in place for a number of processing operations, the edge ring may occasionally be reset; as previously discussed, similar AWC techniques may be performed during such subsequent edge ring placements to compensate for any misalignment between such edge ring and the end effector.

[0143] The techniques discussed herein can also be used to determine the relative offset between two movable components, such as an edge ring and a wafer. For example, Figure 13 A flow chart showing a technique for calibrating wafer placement relative to an edge ring on a wafer support is shown.

[0144] In block 1302, a wafer station of a semiconductor processing tool may be selected. Figure 12 In block 1304, an edge ring may be placed on a wafer support of a selected wafer station and in block 1306, an edge ring may be placed on a wafer support of a selected wafer station using, for example, the method described above for wafer station 1306. Figure 12 The centering technique discussed herein centers the edge ring on the wafer support. Blocks 1304 and 1306 may be optional; the edge ring may also be placed on the wafer support of the selected wafer station via other means, such as manual placement, or the edge ring may be positioned on the wafer support prior to selecting the selected wafer station.

[0145] In block 1308, the wafer handling robot may be caused to retrieve a calibration wafer, such as an unprocessed wafer or a dummy wafer having the same size and thickness as the wafer to be processed, from a storage location. In block 1310, the wafer handling robot may be caused to transfer the calibration wafer to the wafer support of the selected wafer station such that the center of the calibration wafer is formally centered over the center of the edge ring.

[0146] In block 1312, the wafer handling robot can be controlled to retrieve the automatic calibration wafer from a storage location accessible to the wafer handling robot, such as a docking station or other location. In block 1314, the wafer handling robot can be controlled to position the automatic calibration wafer above the wafer support of the selected wafer station such that the automatic calibration wafer is substantially centered above the center point of the calibration wafer and / or the edge ring. In block 1316, the automatic calibration wafer can be controlled so that an edge camera, such as a first camera sensor, of the automatic calibration wafer acquires image data of the gap between the edge ring and the calibration wafer.

[0147] In block 1318, the image data can be analyzed to determine the wafer / edge ring offset between the inner diameter of the edge ring and the outer diameter of the calibration wafer based on the relative gap size in the image. In block 1320, a determination can be made as to whether the wafer / edge ring offset exceeds a predetermined offset threshold. If so, the technique can proceed to block 1322, where the wafer handling robot can be actuated to return the automated calibration wafer to a docking station (or some other temporary holding location), and then to block 1324, where the wafer handling robot can be further actuated to retrieve the calibration wafer from the wafer support of the selected wafer station. For example, lift pins of the wafer support can be caused to lift the calibration wafer off the edge ring so that an end effector of the wafer handling robot can be positioned beneath the calibration wafer. Once so positioned, the lift pins can be further controlled to lower the calibration wafer onto the end effector.

[0148] In block 1326, the wafer handling robot can be actuated to reposition the calibration wafer onto the wafer support of the selected wafer station so that the calibration wafer is centered at a new position that takes into account the wafer / edge ring offset determined in block 1318. The technique can then return to block 1312 to begin further automated calibration wafer imaging of the gap size between the calibration wafer and the edge ring. This repositioning and reanalysis of the calibration wafer's centering with the edge ring can be performed multiple times, for example, until the determined wafer / edge ring offset falls below a predetermined threshold, or until a predetermined number of such repetitions have been performed. If it is determined in block 1320 that the wafer and edge ring are sufficiently centered relative to each other (i.e., the determined wafer / edge ring offset falls within a predetermined threshold), the technique can then proceed to block 1328, where the wafer / edge ring placement calibration can be considered complete. At this point, the calibration wafer can be removed from the wafer station using the wafer handling robot and used to train, for example, an active wafer centering system—in much the same manner as if the calibration wafer had been manually positioned relative to the wafer support, for example, via a fixture or other mechanical centering system. Training an active wafer centering system based on a centered wafer or a calibrated wafer that is otherwise placed onto the end effector of a wafer handling robot is known in the industry and is not described in depth in this disclosure for the sake of brevity.

[0149] It should be understood that the above-described techniques can be implemented in a variety of different ways to achieve similar results. For example, in a tool with multiple wafer handling robots or a wafer handling robot including dual arms / end effectors, one wafer handling robot arm / end effector can be used to place or reposition an object, such as a calibration wafer and / or edge ring, on the wafer support, while another wafer handling robot arm / end effector can be used to support the automated calibration wafer. Thus, for example, a first arm can be used to place an edge ring on the wafer support and then retracted. A second arm can then be used to move the automated calibration wafer over the already placed edge ring to obtain a center-to-center offset measurement between the edge ring and the wafer support. The second arm can then be retracted, and if necessary, the edge ring can be lifted off the wafer support via, for example, lift pins, and the first arm can be used to reposition the edge ring to correct the center-to-center offset between the edge ring and the wafer support. The first arm can then be retracted, and the second arm can again move the automated calibration wafer over the edge ring and wafer support to obtain a second center-to-center offset measurement. This process can be repeated as needed until the desired center-to-center offset between the edge ring and the wafer support is achieved.

[0150] It should also be understood that the automated calibration wafer-guided placement of the wafer and / or edge ring on the wafer support can be an iterative process, wherein an estimate of the relative offset between two structures, such as the wafer and wafer support, the edge ring and wafer support, or the edge ring and wafer, can be obtained using the automated calibration wafer, and the estimate can then be used to guide the repositioning of one of the two structures relative to the other, immobilizable structure. Generally, such automated calibration wafer-assisted placement and evaluation can be repeated until the measured offset falls within a predetermined maximum allowable offset for a given set of components in a given semiconductor processing tool. In some cases where both the edge ring and the calibration wafer are centered using the automated calibration wafer, the relative center offset between any pair of components (edge ​​ring, wafer support, and calibration wafer) that are not directly centered relative to each other can be further evaluated. For example, if the edge ring is centered relative to the wafer support and the calibration wafer is centered relative to the edge ring, the calibration wafer is not directly centered relative to the wafer support (but only indirectly centered via the edge ring centering). In such implementations, the automated calibration wafer can be used to additionally evaluate the centering of the calibration wafer relative to the wafer support. In some such implementations, the predetermined offset thresholds for each pair of structures may be selected so that two of the three center-to-center offsets fall within their respective predetermined offset thresholds, but a third center-to-center offset may actually exceed its respective predetermined offset (of course, the predetermined offset thresholds may be selected so that this does not occur, but in some cases this may allow for unacceptable process uniformity, or may require that some predetermined thresholds be lower than would normally be necessary, which may increase the number of centering iterations that may need to be performed). Generally speaking, in such implementations, the offset thresholds may be selected so as to avoid a two-out-of-three success / one-out-of-three failure scenario for offset threshold compliance in most cases, but if, for example, the edge ring is positioned relative to the wafer support at the limit of a predetermined edge ring / wafer support offset in a particular direction, and the calibration wafer is positioned relative to the edge ring at the limit of a predetermined wafer / edge ring offset in the same direction, the calibration wafer may experience a maximum center-to-center offset from the wafer support that may exceed the maximum center-to-center offset specified for the calibration wafer / wafer support.

[0151] In such implementations, if a two out of three success / one out of three failure scenario occurs, the semiconductor processing tool can take various corrective actions. For example, in some implementations, the controller of the semiconductor processing tool can cause one or both of the calibration wafer and the edge ring to be removed by a wafer handling robot and then repositioned using techniques similar to those discussed above, but using, for example, a more stringent corresponding predetermined offset threshold for placement.

[0152] It should also be understood that some semiconductor processing tools may utilize an automated calibration wafer for both calibration wafer and edge ring placement / centering operations by centering the calibration wafer and edge ring relative to the wafer support (rather than centering the edge ring relative to the wafer support and the calibration wafer relative to the edge ring, or vice versa).

[0153] Once the calibration wafer has been centered on the wafer support or on the edge ring on the wafer support and used to train the active wafer centering system, the trained active wafer centering system can then optionally be tested using an automated calibration wafer to ensure that the trained active wafer centering system can provide reliable centered wafer placement. Figure 14 A flow chart showing such a technique for verifying wafer placement repeatability is shown ( Figure 14 The technique is expected to be implemented after the active wafer centering system has been taught); the technique assumes that the active wafer centering system has been trained using a wafer centered relative to a wafer support, but can also be modified appropriately, such as using a wafer centered relative to an edge ring to train the active wafer centering system.

[0154] In block 1402, a wafer station of a semiconductor processing tool may be selected; the selected wafer station will have a calibration wafer previously centered on its wafer support using the techniques discussed above, and the active wafer centering system associated with the wafer station will have been trained based on the centered position of the calibration wafer. In block 1404, a wafer handling robot of the semiconductor processing tool may be caused by a controller of the semiconductor processing tool to retrieve the calibration wafer from a receiving station, such as a buffer, FOUP, or other location within the semiconductor processing tool. In block 1406, the wafer handling robot may be controlled to place the calibration wafer onto the wafer support of the selected wafer station.

[0155] After the calibration wafer is placed on the wafer support, a wafer handling robot may be caused to retrieve the automated calibration wafer from, for example, a docking station or other storage location in block 1408. In block 1410, the wafer handling robot may be actuated to position the automated calibration wafer over the calibration wafer and wafer support of the selected wafer station such that the first camera sensor of each automated calibration wafer has the edge of the calibration wafer and the edge of the wafer support within its field of view.

[0156] In block 1412, an automated calibration wafer may be used to obtain image data of the gap between the edge of the calibration wafer and the edge of the wafer support. A determination may be made in block 1414 regarding the offset between the center of the wafer support and the center of the calibration wafer; this offset may be stored for later reference. In block 1416, a count X may be incremented to X+1, and in block 1418, a determination may be made as to whether X exceeds a predetermined threshold Y. X may represent the number of test placements performed as part of the technique, while Y may represent the total number of test placements to be performed as part of the technique.

[0157] If, as determined in block 1418, X is not greater than Y, the technique may proceed to block 1420 before returning to block 1404. In block 1420, the calibration wafer may be returned to its original position in the holding station, or to another position with a randomized offset, before the wafer handling robot returns to its default or "home" position in block 1422. The randomized offset may be selected to fall within the typical expected offset of wafers under normal operational use, e.g., an offset of less than 0.8 mm. Thus, when the wafer handling robot retrieves the calibration wafer again in block 1404 (which typically results in returning to the same position each time to retrieve the calibration wafer), the calibration wafer will have a corresponding randomized offset position relative to the end effector of the wafer handling robot. It should also be understood that randomization of wafer placement may occur at other times, such as just prior to retrieving the calibration wafer from the holding station or other location, and that the wafer handling robot may be subjected to random placement to similarly randomize the position of the calibration wafer relative to the end effector. Such randomization may be used to account for slight misalignments of wafers placed within the holding station during normal operation.

[0158] If it is determined in block 1418 that sufficient test wafer placement has been performed, the technique may proceed to block 1424, where the center offset of the Y wafer placement may be evaluated or analyzed. Such analysis may include any of a variety of different analytical techniques or tests, such as determining statistical parameters of a group of captured test wafer offsets and comparing them to corresponding thresholds. For example, the mean, median, and standard deviation of the offsets may be determined and compared to corresponding thresholds for such values ​​to determine whether the wafer placement resulting from the test technique is acceptable in consistency. In block 1426, the offsets may be compared to these thresholds (or, more precisely, the statistical parameters derived from the offsets may be compared to their corresponding thresholds) to determine whether the test was successful. If the comparison in block 1426 indicates that one or more allowed parameters were exceeded, the technique may proceed to block 1428, where an error condition may be generated. If the comparison in block 1426 indicates that one or more parameters are within acceptable limits, the technique may proceed to block 1430, where the technique may be successfully completed.

[0159] It should be understood that similar techniques can also be implemented using edge rings, such as performing repeated edge ring placements and randomizing the end effector / edge ring offset between each placement to evaluate the repeatability of the edge ring placement.

[0160] Figure 15 A flow chart of a technique for evaluating the height of an edge ring is shown. In block 1502, a wafer station of a semiconductor processing tool may be selected; the selected wafer station should already have an edge ring located on its wafer support. For example, an edge ring may be left in place on the wafer support during processing operations, and an edge ring may be performed at regular intervals during such operations. Figure 15 A technique is provided to determine whether an edge ring has degraded in a non-uniform manner (or to an unacceptable degree, regardless of uniformity) due to exposure to repeated semiconductor wafer processing cycles.

[0161] The automatic calibration wafer may be retrieved by the wafer handling robot and then transferred to a wafer support of the selected wafer station in block 1504. Such transfer of the automatic calibration wafer to the wafer support may include, for example, lifting the calibration wafer off the wafer handling robot via lift pins and then lowering the automatic calibration wafer onto an edge ring by retracting the lift pins, placing the automatic calibration wafer directly onto the edge ring.

[0162] In block 1508, the auto-calibration wafer may be caused to obtain distance measurements between the auto-calibration wafer and the wafer support using proximity sensors of the auto-calibration wafer. In some implementations, such distance measurements may be obtained using proximity sensors located at at least three locations about the circumference of the auto-calibration wafer, thereby determining the orientation of a plane defined by the auto-calibration wafer relative to a plane defined by the upper surface of the wafer support. If the distance between the two planes at any point about a circle centered on the diameter of the edge ring at or near the inner diameter of the edge ring exceeds a certain threshold, this may indicate that the thickness of the edge ring is out of tolerance and the edge ring should be replaced.

[0163] In block 1510, the autocalibration wafer can be removed from the wafer support using a wafer handling robot. In block 1512, the obtained distance measurements can be evaluated and a determination can be made as to whether the measured distances indicate that the edge ring falls within acceptable height limits. For example, if any of the proximity distances are below (or above) a predetermined threshold, this can indicate that the edge ring height is too small (or too large). Another metric that can be used to assess edge ring height is the difference between different distance measurements. For example, for a given measurement cycle using the autocalibration wafer, the difference between the maximum distance measurement for the edge ring and the shortest distance measurement for the edge ring can be compared to another predetermined threshold to determine whether the edge ring height around the edge ring circumference has varied by an unacceptable amount. If it is determined in block 1512 that the distance measurements fall within acceptable limits, the technique can proceed to block 1516, where a success condition can be determined. If it is determined in block 1512 that the distance measurements do not fall within acceptable limits, the technique can proceed to block 1514, where a failure or error condition can be generated. Such a condition may cause the semiconductor processing tool to halt further processing operations at the wafer station until a new edge ring has been installed, centered, and a height measurement has been taken.

[0164] As previously mentioned, in some implementations, an automatic calibration wafer may be used to determine dynamic characteristics of a semiconductor processing tool, such as vibration and tilt of lift pins. Figure 16 A flow chart showing a technique for evaluating lift pin vibration.

[0165] Figure 16 The technique may begin at block 1602, where a wafer station for lift pin vibration evaluation may be selected. In block 1604, a wafer handling robot may be actuated to retrieve an autocalibration wafer from, for example, a storage location (e.g., a docking station). In block 1606, the autocalibration wafer may be positioned above a wafer support of the selected wafer station, and the lift pins of the wafer support may then be caused to lift the autocalibration wafer off the end effector of the wafer handling robot. In block 1608, the autocalibration wafer may begin acquiring vibration data from its vibration sensor; it will be understood that in some implementations, such data may be acquired earlier or continuously. In block 1610, the lift pins may be actuated to move the autocalibration wafer, for example, vertically relative to the wafer support. In some implementations, such movement may involve lowering the autocalibration wafer onto the wafer support and then raising it again, as a semiconductor wafer would typically undergo during normal wafer placement operations. In other implementations, the lift pins may be actuated in a manner that does not correspond to normal wafer placement movement but is instead designed to more readily induce a specific vibration response. In either case, the autocalibration wafer is exposed to movement via actuation of the lift pins, during which vibration data can be collected by one or more vibration sensors.

[0166] In block 1612, the wafer handling robot can be controlled to retrieve the automated calibration wafer from the lift pins, and in block 1614, the vibration data can be analyzed to determine whether it falls within acceptable limits. For example, if the amplitude of the vibration exceeds a predetermined threshold, or if the amplitude of a particular frequency component of the vibration exceeds a predetermined threshold, the vibration measurement can be determined to be outside acceptable limits, and the technique can proceed to block 1618, where an error condition is generated. The duration of the lift pin movement during each phase of the movement can also be measured and compared to an acceptable range to determine whether an error condition exists in the lift pin mechanism. In such a case, the semiconductor processing tool can, for example, suspend wafer processing operations using the wafer processing station until the lift pin mechanism is serviced and the issue is resolved. If the lift pin mechanism is controlled by an actuator, a closed-loop system can be implemented to automatically calibrate the lift pin mechanism using the vibration data. In some implementations, the semiconductor processing tool can generate a warning indicating that the lift pin mechanism requires maintenance or repair, but can continue semiconductor processing operations at the wafer station (and using the lift pin mechanism) until a subsequent vibration evaluation of the lift pins indicates that vibration occurring during lift pin actuation has exceeded a second set of acceptable limits. In some such implementations, after encountering an error condition, the semiconductor processing tool may operate the lift pin mechanism at a lower performance level (e.g., at a lower speed than normal lift pin mechanism speed) to potentially reduce the magnitude of vibration experienced. This lower capacity operation may continue until the lift pin mechanism is repaired and the issue is resolved, or until the vibration generated during operation of the lift pin mechanism has degraded to further unacceptable levels, in which case the wafer station and lift pin mechanism may be temporarily suspended until maintenance can be performed. Vibration data from the lift pin mechanism may be sent to an off-tool data center with big data and machine learning capabilities. The data center may receive vibration data from a large number of similar semiconductor processing tools to develop a healthy versus unhealthy lift pin vibration signature. In such a case, either the raw vibration data or the vibration signature (after machine learning feature extraction) may be sent to the data center. The data center may use the received vibration data from a batch of semiconductor processing tools to train a machine learning model, such as a neural network, TensorFlow, etc., to classify a lift pin mechanism as healthy or unhealthy.

[0167] Another testing technique that can be performed with an automated calibration wafer is to evaluate the levelness of the wafer support (or other equipment such as the end effector of a wafer handling robot, lift pin mechanism, load port module (LPM), wafer support, ESC, etc.). Figure 17 A flow chart is shown for a technique for assessing the levelness of a wafer support, but the technique can be implemented for a variety of different wafer handling components.

[0168] In block 1702, a wafer station of a semiconductor processing tool may be selected for levelness measurement of a wafer support.

[0169] In block 1704, a wafer handling robot of the semiconductor processing tool may be caused to retrieve the auto-calibration wafer from a docking station or other storage location, and the wafer handling robot may be controlled to place the auto-calibration wafer directly onto the wafer support in block 1706. If an edge ring is already present on the wafer support and would interfere with the auto-calibration wafer being placed directly onto the wafer support, the edge ring may be removed by the wafer handling robot of the semiconductor processing tool before placing the auto-calibration wafer onto the wafer support.

[0170] Once the autocalibration wafer is placed on the wafer support, a levelness measurement of the autocalibration wafer may be obtained using, for example, an orientation sensor (e.g., an accelerometer or tilt sensor) in block 1708. In some implementations, multiple levelness sensors may be used to obtain such measurements.

[0171] In block 1710, the automated calibration wafer may be retrieved from the wafer support using a wafer handling robot, and in block 1712, a determination may be made as to whether the levelness measurement of the wafer support falls within acceptable limits. If not, the technique may proceed to block 1714, where an error condition may be generated. If so, the technique may proceed to block 1716, where a success condition may be generated.

[0172] It should be understood that the various techniques described herein can be combined in a variety of ways to provide a fully automated system for configuring a semiconductor processing tool. For example, a semiconductor processing tool can be configured with an "initial setup" mode that can be used in the following situations: the tool can retrieve an edge ring and center the edge ring on each wafer support in a semiconductor processing chamber using an automated calibration wafer, then center the calibration wafer and associated edge ring for each wafer support using an automated calibration wafer, train an active wafer centering system and a wafer handling robot using the centered calibration wafer, and then verify that the trained active wafer centering system produces reliable wafer placement. The semiconductor processing tool can also periodically perform various health checks, such as checking whether the center-to-center offset of the edge ring and the calibration wafer has drifted to an unacceptable distance, checking whether the edge ring height remains within an acceptable range after a predetermined period of time or number of wafer processing operations, and / or checking whether the wafer support is level and / or whether lift pin vibration is within an acceptable range.

[0173] As described above, a controller can be part of a system that can include semiconductor processing equipment, which includes one or more processing tools, one or more chambers, one or more platforms for processing, and / or specific processing components (wafer pedestals, gas flow systems, etc.). These systems can be integrated with electronic devices for controlling their operation before, during, and after processing of semiconductor wafers or substrates. The electronic devices can be referred to as "controllers" and can control various components or subcomponents of one or more systems. Depending on the processing requirements and / or the type of system, the controller can be programmed to control any of the processes disclosed herein, as well as various parameters that affect semiconductor processing, such as the delivery of process gases, temperature settings (e.g., heating and / or cooling), pressure settings, vacuum settings, power settings, radio frequency (RF) generator settings, RF matching circuit settings, frequency settings, flow rate settings, fluid delivery settings, position and operation settings, wafer transfer in and out of tools and other transfer tools and / or load locks connected to or interfaced with a specific system.

[0174] In general, a controller can be defined as an electronic device having various integrated circuits, logic, memory, and / or software that receive instructions, issue instructions, control operations, enable cleaning operations, enable endpoint measurements, and the like. The integrated circuits can include chips in the form of firmware that stores program instructions, digital signal processors (DSPs), chips defined as application specific integrated circuits (ASICs), and / or one or more microprocessors or microcontrollers that execute program instructions (e.g., software). The program instructions can be instructions sent to the controller in the form of various separate settings (or program files) that define operating parameters for performing specific processes on or for a semiconductor wafer or system. In some embodiments, the operating parameters can be part of a recipe defined by a process engineer to complete one or more process steps during the manufacture of one or more layers, materials, metals, oxides, silicon, silicon dioxide, surfaces, circuits, and / or the die of the wafer.

[0175] In some implementations, the controller can be part of or coupled to a computer that is integrated with the system, coupled to the system, otherwise networked to the system, or a combination thereof. For example, the controller can be in the "cloud" or all or part of a wafer fab host system that can allow remote access to wafer processing. The computer can enable remote access to the system to monitor the current progress of manufacturing operations, review the history of past manufacturing operations, review trends or performance metrics for multiple manufacturing operations, change parameters of the current process, set processing steps to follow the current process, or start a new process. In some examples, a remote computer (e.g., a server) can provide process recipes to the system via a network (which can include a local network or the Internet). The remote computer can include a user interface that enables the input or programming of parameters and / or settings, which are then sent from the remote computer to the system. In some examples, the controller receives instructions in the form of data that specify parameters for each processing step to be performed during one or more operations. It should be understood that the parameters can be specific to the type of process to be performed and the type of tool the controller is configured to interface with or control. Thus, as described above, the controller can be distributed, for example, by including one or more discrete controllers networked together and working toward a common purpose (e.g., processing and control as described herein). An example of a distributed controller for such a purpose is one or more integrated circuits on a chamber communicating with one or more integrated circuits remotely (e.g., at a platform level or as part of a remote computer), which combine to control processing on the chamber.

[0176] Exemplary systems may include, but are not limited to, plasma etch chambers or modules, deposition chambers or modules, spin rinse chambers or modules, metal plating chambers or modules, cleaning chambers or modules, bevel edge etch chambers or modules, physical vapor deposition (PVD) chambers or modules, chemical vapor deposition (CVD) chambers or modules, atomic layer deposition (ALD) chambers or modules, atomic layer etch (ALE) chambers or modules, ion implantation chambers or modules, track chambers or modules, and any other semiconductor processing system that may be associated with or used in the manufacture and / or preparation of semiconductor wafers.

[0177] As described above, depending on one or more processing steps to be performed by the tool, the controller can communicate with one or more other tool circuits or modules, other tool components, cluster tools, other tool interfaces, adjacent tools, neighboring tools, tools located throughout the factory, a host computer, another controller, or tools used in the material transport of wafer containers to and from tool locations and / or load ports in a semiconductor manufacturing facility.

[0178] It should be understood that while the above discussion focuses on auto-calibration wafer systems that typically feature multiple camera sensors (with one or more additional sensors in various other implementations), some implementations may feature a single centrally mounted camera sensor and no edge-mounted camera sensors, or only edge-mounted camera sensors and no centrally mounted camera sensors; in some cases, such implementations may provide less functionality than implementations with both centrally mounted and edge-mounted camera sensors, but this is not necessarily the case. The present disclosure should also be understood to encompass such alternative implementations.

[0179] As used herein, the term "wafer" may refer to a semiconductor wafer or substrate or other similar types of wafers or substrates. As used herein, a wafer station may refer to any location in a semiconductor processing tool where a wafer may be placed during any of a variety of wafer handling operations or wafer transfer operations. A wafer support is used herein to refer to any structure in a wafer station configured to receive and support a semiconductor wafer, such as a pedestal, electrostatic chuck, wafer support rack, etc.

[0180] As used herein, the term "nominal centering" refers to the relative placement of two or more objects so that some locations, such as center points or similar locations, are approximately aligned with each other in the XY plane. This alignment may not be perfect due to various reasons, such as slippage of one of the objects, sensor drift, etc., but in most cases, nominally centered objects can be within one or two millimeters of ideally centered objects.

[0181] It should also be understood that the use of ordinal numbers herein, such as (a), (b), (c), ..., is for organizational purposes only and is not intended to convey any particular order or importance to the items associated with each ordinal number. For example, "(a) obtaining information related to speed; and (b) obtaining information related to position" should include obtaining information related to position before obtaining information related to speed, obtaining information related to speed before obtaining information related to position, and obtaining information related to position while obtaining information related to speed. Nevertheless, there may be situations where certain items associated with ordinal numbers may require a specific order, for example, "(a) obtaining information related to speed, (b) determining acceleration based on the information related to speed, and (c) obtaining information related to position"; in this example, (a) needs to be performed before (b) because (b) depends on the information obtained in (a), but (c) can be performed before or after either (a) or (b).

[0182] It should be understood that phrases such as "for each <item> of the one or more <items>" or "of each <item>"<item>)" (if used herein) should be understood to include both single groups of items and multiple groups of items, that is, the phrase "for...each" is used in the sense that it is used in a programming language to refer to each item in the entire group of items being referred to. For example, if the group of items being referred to is a single item, then "each" will refer only to that single item (despite the fact that the dictionary definition of "each" is often defined as meaning "each of two or more things"), and does not mean that there must be at least two of those items. Similarly, when a selected item can have one or more sub-items and a selection is made of one of those sub-items, it should be understood that, where the selected item has one and only one sub-item, selecting that one sub-item is inherently selecting the item itself.

[0183] It should also be understood that reference to a plurality of controllers generally configured to perform various functions is intended to encompass situations where only one of the controllers is configured to perform all of the functions disclosed or discussed, as well as situations where various controllers each perform a subset of the functions discussed. For example, an auto-calibration wafer may include a controller configured to control the operation of various sensors on the auto-calibration wafer and communicate data from the sensors to another controller associated with a semiconductor processing tool; the semiconductor processing tool controller may then analyze such data to determine various operating parameters for use with the semiconductor processing tool.

[0184] Various modifications to the embodiments described in this disclosure will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments without departing from the spirit or scope of this disclosure. Therefore, the claims are not intended to be limited to the embodiments presented herein, but should be accorded the widest scope consistent with this disclosure, principles, and novel features disclosed herein.

[0185] Certain features of this specification described in the context of separate implementations may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually in multiple embodiments or in any suitable subcombination. Furthermore, while features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination may in some cases be omitted from that combination, and a claimed combination may be directed to subcombinations or variations of subcombinations.

[0186] Similarly, although operations are depicted in a particular order in the accompanying drawings, this should not be understood as requiring these operations to be performed in the particular order shown or in a continuous order, or to perform all the operations shown to achieve the desired result. In addition, the accompanying drawings can schematically depict another example process in the form of a flow chart. However, other operations that are not depicted can be combined in the example process schematically shown. For example, one or more additional operations can be performed before, after, simultaneously or between any of the illustrated operations. In some cases, multitasking and parallel processing may be advantageous. In addition, the separation of the various system components in the above-described embodiments should not be understood as requiring such separation in all embodiments, and should be understood as that the program components and system can be generally integrated together in a single software product or packaged into multiple software products. In addition, other embodiments all fall within the scope of the following claims. In some cases, the actions described in the claims can be performed in different orders and still achieve the desired result.< / item>

Claims

1. A system for assisting in calibration of a wafer handling robot of a semiconductor processing tool, comprising: An automatic calibration chip comprising: a substrate sized to be transported by the wafer handling robot and having a first side configured to contact an end effector of the wafer handling robot when the substrate is being transported by the wafer handling robot; a plurality of first imaging sensors supported by the substrate and positioned at a plurality of locations offset from a common point on the substrate, each of the first imaging sensors having a downwardly facing field of view when the substrate is positioned with the first side facing downward; as well as A first controller is communicatively coupled to each of the first imaging sensors. 2 . The system of claim 1 , wherein the first imaging sensors are arranged in a circular array around the common point.

3. The system of claim 1, wherein the substrate is nominally circular and has the same diameter as a semiconductor wafer that the semiconductor processing tool is configured to process.

4. The system of claim 1, wherein the substrate is nominally circular and has the same diameter as an edge ring with which the semiconductor processing tool is configured to be used.

5. The system of claim 1, wherein the substrate is nominally circular and has a diameter between an outer diameter and an inner diameter of an edge ring with which the semiconductor processing tool is configured for use.

6. The system of claim 1 , wherein the substrate is nominally round and has a diameter that is within ±10% of a mean value between an outer diameter and an inner diameter of an edge ring, the semiconductor processing tool being configured to use the edge ring.

7. The system of claim 1, wherein the substrate is a nominal circular disk having a diameter selected from the group consisting of 200 mm, 300 mm, and 450 mm. 8 . The system of claim 1 , wherein the auto-calibration wafer further comprises a power supply configured to provide power to at least the first controller and the first imaging sensor.

9. The system of claim 8, wherein: The power source is a rechargeable battery, and The auto-calibration chip also includes a wireless charging feature configured to charge the rechargeable battery when the rechargeable battery is docked with an electromagnetic field.

10. The system of claim 1, wherein: The automatic calibration chip also includes a first wireless communication interface, and The first wireless communication interface is in communication with the first controller.

Images