Method for detecting wafer slip

Abstract

An exemplary method for detecting substrate slip includes causing a first sensor and a second sensor of an in-situ monitoring system to sweep over a substrate when the substrate is subjected to polishing with a rotatable platform. A first sequence of signal values from the first sensor and a second sequence of signal values from the second sensor include signal strengths relative to the thickness of the layer. For each signal value of at least some of the first sequence of signal values and the second sequence of signal values, the method may include determining a first location and a second location on the substrate, respectively. The method may also include initiating a slip alarm if the signal strength varies by 30% or more from the first sequence of signal values to the second sequence of signal values, the location of the second signal value on the substrate cannot be determined, or a combination thereof.

Description

Technical Field

[0001] This application claims the benefit and priority of U.S. Patent Application No. 18 / 185,849, filed March 17, 2023, entitled “Method for Detecting Wafer Slippage,” which is incorporated herein by reference in its entirety.

[0002] This technology relates to the monitoring of wafer slip during chemical mechanical polishing (CMP). More specifically, this technology relates to a non-optical method for monitoring wafer slip during CMP. Background Technology

[0003] Integrated circuits are typically formed on a substrate by sequentially depositing conductive, semiconducting, or insulating layers on a silicon wafer. Various manufacturing techniques require planarization of the layers on the substrate. For example, one manufacturing step involves depositing a conductive fill layer on a patterned insulating layer to fill trenches or holes in the insulating layer. The fill layer is then polished, a process known as planarization, until the raised patterns of the insulating layer are exposed. After planarization, the portions of the conductive fill layer retained between the raised patterns of the insulating layer form vias, plugs, and traces that provide conductive paths between thin-film circuits on the substrate. Furthermore, planarization can be used to planarize dielectric layers used in lithography. However, the necessary liquids and components used during planarization often prevent optical sensors from detecting substrate slippage.

[0004] Therefore, there is a need for improved sensors and methods that can quickly alert to substrate slippage events, even when using abrasive liquids. This technology addresses these and other needs. Summary of the Invention

[0005] An exemplary method for detecting substrate slippage includes contacting a surface of a layer of the substrate with a polishing pad. This method may include inducing relative movement between the substrate and the polishing pad. This method may include having a first sensor and a second sensor of a plurality of sensors in an in-situ monitoring system sweep across the substrate while the layer of the substrate is being polished using a rotatable platform. This method may include generating a first sequence of signal values ​​from the first sensor and a second sequence of signal values ​​from the second sensor from the in-situ monitoring system. The first and second signal value sequences include signal strength relative to the thickness of the layer. This method may include detecting the time it takes for the first sensor to traverse the leading edge and the trailing edge of the substrate from the first signal value sequence from the first sensor. This method may also include detecting a second time it takes for the second sensor to traverse the leading edge and the trailing edge of the substrate from the second signal value sequence from the second sensor. For each signal value in at least some of the first signal value sequences, this method may include determining a position on the substrate. The first signal value may be based on the time it takes for the first sensor to cross the leading edge of the substrate and the time it takes for the first sensor to cross the trailing edge of the substrate. For each signal value in at least some of the second signal value sequences, the method may include determining the position on the substrate. The second signal value may be based on the time it takes for the second sensor to cross the leading edge of the substrate and the time it takes for the second sensor to cross the trailing edge of the substrate. The method may also include activating a slip alarm if the signal strength changes by 30% or more from the first signal value sequence to the second signal value sequence, making it impossible to determine the position of the second signal value on the substrate, or a combination thereof.

[0006] In some embodiments, determining this position may include determining the first derivative of the first signal value sequence and identifying valleys and peaks in the first derivative. The method may also include a third signal value sequence from a third sensor among these sensors. This third sensor may be positioned between the first and second sensors. The method may include activating a misalignment alarm only if the signal strength changes by 30% or more from both the first signal value sequence to the third signal value sequence and from the third signal value sequence to the second signal value sequence, making it impossible to determine the position of the second and third signal values ​​on the substrate, or a combination thereof. The method may include stopping the relative movement upon activation of the misalignment alarm. The method may also include stopping the relative movement by a controller. In one embodiment, detecting the signal value sequence includes detecting conductive layers within the leading and trailing edges of the substrate.

[0007] An exemplary method for detecting substrate slippage may include contacting the surface of a layer of the substrate with a polishing pad. This method may include inducing relative movement between the substrate and the polishing pad. This method may include monitoring the layer of the substrate using an in-situ monitoring system to generate a signal, wherein the signal strength depends on the thickness of the layer. This method may include filtering the signal to generate a filtered signal. This method may also include triggering a misalignment alarm if a second signal strength value differs from a first signal strength value by 30% or more.

[0008] In some embodiments, this method may include an in-situ monitoring system with multiple sensors, wherein these sensors generate the signal. The method may include positioning the sensors to intermittently scan beneath the substrate, such that at least one sensor is positioned beneath the substrate at any given time. In one embodiment, at least one first sensor is positioned in a platform located beneath the polishing pad. The method may include stopping the relative movement if the slip alarm is activated. The method may also include stopping the relative movement by a controller. The method may include at least a first sensor, a second sensor, and a third sensor. In some embodiments, the slip alarm is activated only if the signal strength changes by 30% or more from both the first signal strength value to the second signal strength value and the second signal strength value to the third signal strength value.

[0009] This technology may also relate to exemplary computer programs tangibly encoded on a non-transitory computer-readable medium. Such computer programs may include instructions to cause a computer system to receive a first sequence of signal values ​​from a first sensor of an in-situ monitoring system. These instructions may also cause a second sequence of signal values ​​to be received from a second sensor. The first and second sequence of signal values ​​may include signal strengths relative to the thickness of the layer. The first and second sensors may sweep across and monitor the substrate during polishing. These instructions may cause the computer system to detect, from the first sequence of signal values, the time it takes for the first sensor to cross the leading edge of the substrate and the time it takes for the first sensor to cross the trailing edge of the substrate. These instructions may also cause the computer system to detect, from the second sequence of signal values, the time it takes for the second sensor to cross the leading edge of the substrate and the time it takes for the second sensor to cross the trailing edge of the substrate. For each signal value in at least some of the first sequence of signal values, the position on the substrate may be determined based on the time it takes for the first sensor to cross the leading edge of the substrate and the time it takes for the first sensor to cross the trailing edge of the substrate. For each signal value in at least some of the second signal value sequences, the position on the substrate can be determined based on the time it takes for the second sensor to cross the leading edge of the substrate and the time it takes for the second sensor to cross the trailing edge of the substrate. If the signal strength changes by 30% or more from the first signal value sequence to the second signal value sequence, making it impossible to determine the position of the second signal value on the substrate, or a combination thereof, then such instructions may also cause the computer to activate a slip alarm.

[0010] In some embodiments, activating this slip alarm can cause the grinding to stop. These instructions may also include the computer system automatically stopping the grinding process. These instructions may cause the computer system to receive a third signal value sequence from a third sensor of the in-situ monitoring system. This third signal value sequence may include the signal strength relative to the thickness of the layer. The third sensor may be circumferentially disposed between the first sensor and the second sensor. In embodiments, the slip alarm may be activated only if the signal strength changes by 30% or more from both the first signal value sequence to the third signal value sequence and from the third signal value sequence to the second signal value sequence, making it impossible to determine the position of the second signal value and the third signal value on the substrate, or a combination thereof. In some embodiments, the instructions for determining the position of the first signal value on the substrate include instructions for determining the distance between the center of the bearing head holding the substrate and the axis of rotation of the rotatable platform. These determining instructions may be based on the time it takes for the first sensor to cross the leading edge and the trailing edge. These instructions may also be based on this distance to determine the position of the first signal value on the substrate.

[0011] Compared to conventional systems and methods, this technique offers numerous advantages. For example, embodiments of this technique allow for the automatic termination of the polishing process, preventing damage to the substrate and the polishing system. Furthermore, the systems and methods discussed herein can operate under all polishing conditions and can be used to provide the necessary redundancy for high-speed and high-quality slip evaluation. These and other embodiments, along with their many advantages and features, are described in more detail below with reference to the accompanying drawings. Attached Figure Description

[0012] A further understanding of the nature and advantages of the disclosed technology can be achieved by referring to the remainder of the specification and the accompanying drawings.

[0013] Figure 1 This is a schematic cross-sectional side view of a chemical mechanical grinding system, including a current monitoring system.

[0014] Figure 2 This is a schematic top view of a chemical mechanical grinding system.

[0015] Figure 3 This is a schematic circuit diagram of the drive system used in an electromagnetic induction monitoring system.

[0016] Figures 4A to 4C A schematic diagram showing the progress of substrate grinding is shown.

[0017] Figure 5 An illustrative diagram showing an idealized signal from an electromagnetic induction monitoring system.

[0018] Figure 6 An illustrative diagram showing the raw and filtered signals from the electromagnetic induction monitoring system.

[0019] Figure 7 An illustrative diagram showing the raw and filtered signals from the electromagnetic induction monitoring system.

[0020] Figure 8A The graph shows the signal from the current monitoring system as the core scans on the substrate.

[0021] Figure 8B The graph shows the first derivative of the signal.

[0022] Figure 8C This is a diagram showing the expansion of the first derivative of a portion of the signal from the leading edge of the substrate.

[0023] Figure 8D The diagram shows the expansion of the first derivative of a portion of the signal from the leading edge of the support ring.

[0024] Figure 8E This is a diagram showing the expansion of the first derivative of a portion of the signal from the trailing edge of the wafer.

[0025] Figure 8F The diagram shows the expansion of the first derivative of a portion of the signal from the trailing edge of the support ring.

[0026] Figure 9 This is a schematic diagram illustrating the process used to calculate the radial position of a measurement.

[0027] Figure 10 Includes a graph of sensor readings (%) based on wafer position (mm), and shows a graphical user interface that will be displayed by the controller.

[0028] Figure 11 A schematic diagram illustrating the calculation of the measurement location (based on the radial distance from the center of the substrate).

[0029] Several of the figures are included as illustrative. It should be understood that these figures are for illustrative purposes and should not be considered to be to scale unless specifically stated otherwise. Furthermore, these figures are provided as illustrative to aid understanding and may not include all aspects or information compared to a realistic representation, and may include exaggerated data for illustrative purposes.

[0030] In the accompanying drawings, similar parts and / or features may have the same reference numerals. Furthermore, various parts of the same type may be distinguished by adding a letter after the reference numeral to differentiate them. If only the first reference numeral is used in the specification, this specification applies to any of the similar parts having the same first reference numeral, regardless of the letter used. Detailed Implementation

[0031] Substrate processing may involve the formation and removal of material to produce any number of structures. During processing, particle generation and contamination can affect the compatibility of interface layers and the quality of layer formation. Chemical-mechanical polishing (CMP) or any other type of polishing or removal can be performed to produce relatively flat surfaces, such as within system tolerances. CMP and other processing operations are commonly used in semiconductor processing to planarize or polish layers of material formed on semiconductor substrates. Conventional CMP polishing involves positioning the substrate face down on a polishing pad, with a carrier holding the substrate against the rotating polishing pad. Typically, an abrasive liquid, such as a slurry containing abrasive particles, is supplied to the surface of the polishing pad during rotation. Material formed along the substrate is removed through a combination of chemical interactions of the abrasive slurry and mechanical interactions with the polishing pad.

[0032] As can be clearly seen from the above, during CMP planarization, the substrate is not physically connected to the polishing pad or the bearing head. Instead, the substrate relies on sufficient force between the polishing pad and the bearing head to hold it in place during polishing. If processing conditions are incorrect, or if a sudden change occurs, the substrate may slip or be pushed out of the polishing head. Currently, optical sensors are used to detect slippage (e.g., undesirable movement where the substrate moves at least partially or completely from under the bearing head). However, optical sensors mounted solely on the bearing head have proven insufficient for in-situ monitoring. For example, due to part orientation, the optical sensor may only be mounted on the trailing edge of the bearing head. Therefore, if slippage occurs elsewhere around the bearing head, it will not be detected. Similarly, optical sensors detect slippage based on light reflection from the substrate. However, polishing liquids, especially opaque polishing slurries, may cover the upper surface of the slipping substrate, hindering optical detection.

[0033] This technology overcomes these challenges, enabling rapid and accurate detection of substrate slippage beneath the carrier. Specifically, it utilizes a unique array of electromagnetic sensors positioned within the platform. This array allows for the automatic recording of substrate slippage based on substrate thickness, edge detection, or both. The detection method used with these electromagnetic sensors is highly sensitive, even when polishing liquids or polishing components interfere with optical sensors. Furthermore, the electromagnetic detection method can record slippage events within very short timeframes, such as even less than a single polishing rotation. This short alarm cycle provides an additional advantage, as continuing polishing after slippage could damage the wafer, polishing head, or a combination thereof.

[0034] While the remainder of the disclosure will conventionally identify specific film polishing processes utilizing the disclosed techniques, it will be readily understood that such systems and methods are equally applicable to a wide variety of other semiconductor processing operations and systems. Therefore, this technique should not be considered so limited as to be used only with the described polishing systems or processes. Before describing systems, methods, or operations with exemplary process sequences according to some embodiments of this technique, this disclosure will discuss a possible system that can be used with this technique. It should be understood that this technique is not limited to the described equipment, and the discussed processes can be performed in any number of processing chambers and systems, and can be modified in any number of ways, some of which will be described below.

[0035] Figure 1An example of a grinding system 100 for a chemical mechanical grinding apparatus is shown. The grinding system 100 includes a rotatable disc-shaped platform 102 on which a grinding pad 104 is located. The platform 102 is operable to rotate about its central axis 106. For example, a motor 108 can rotate a drive shaft 110 to rotate the platform 102. The grinding pad 104 may be a two-layer grinding pad having an outer layer 112 and a softer backing layer 114. The grinding system 100 may include a supply port or a combined supply flushing arm 116 to dispense a grinding liquid 118, such as a slurry, onto the grinding pad 104.

[0036] The bearing head 128 is operable to hold the substrate 130 against the polishing pad 104. The bearing head 128 is suspended on a support structure 132 (e.g., a turntable or track) and connected to a bearing head rotation motor 136 by a drive shaft 134, allowing the bearing head to rotate about axis 138. Depending on the situation, the bearing head 128 can oscillate laterally, for example, on a slider on the turntable or track 132; or through rotational oscillation of the turntable itself. In operation, the platform 102 rotates about its central axis 106, and the bearing head 128 rotates about its central axis 138 and translates laterally across the top surface of the polishing pad 104. In the presence of multiple bearing heads, each bearing head 128 can independently control its polishing parameters; for example, each bearing head can independently control the pressure applied to each respective substrate. The carrier head 128 may include a flexible diaphragm 140 and a plurality of pressurizable chambers 142. The flexible diaphragm has a substrate mounting surface that contacts the back side of the substrate 130. The pressurizable chambers apply different pressures to different areas of the substrate 130, such as different radial areas. The carrier head may also include a support ring 144 to hold the substrate.

[0037] In some implementations, the polishing system 100 includes an in-situ monitoring system 146 that generates signals representing the thickness of one or more layers on the substrate 130 being polished. For example, the in-situ monitoring system 146 may be an optical monitoring system, such as a spectral monitoring system or an eddy current monitoring system. However, as described above, in some embodiments, the monitoring system 146 discussed herein is an electromagnetic current monitoring system. The in-situ monitoring system 146 may be coupled to a controller 148 that can detect the polishing endpoint, adjust polishing parameters based on measurements to reduce polishing non-uniformity, or detect slippage of the substrate 130. At least some sensor components of the in-situ monitoring system 146 (discussed in more detail below) may be located in recesses 150 formed in the platform 102.

[0038] The polishing system 100 includes an in-situ monitoring system 146 that generates signals representing the thickness of one or more polishing system 100 components. Specifically, the in-situ monitoring system 146 may be an electromagnetic induction monitoring system. This electromagnetic induction monitoring system can operate by generating eddy currents in a conductive layer or by generating current in a conductive ring. In operation, the polishing system 100 can use the monitoring system 146 to determine the thickness of one or more polishing system 100 components (e.g., one or more layers of substrate 130).

[0039] Monitoring system 146 may include a sensor 154 mounted in a recess 150 of platform 102. Sensor 154 may include a magnetic core 156 at least partially positioned in the recess 150 and at least one coil 158 wound around the magnetic core 156. Drive and sensing circuitry 160 is electrically connected to coil 158. Drive and sensing circuitry 160 generates signals that can be sent to controller 148. Although illustrated external to platform 102, some or all of drive and sensing circuitry 160 may be mounted within platform 102. Rotary coupler 162 may be used to electrically connect components (e.g., coil 158) in rotatable platform 102 to components (e.g., drive and sensing circuitry 160) external to platform 102.

[0040] Depending on the situation, a recess 164 may be formed at the bottom of the polishing pad 104 covering the recess 150. Depending on the situation, a portion of the magnetic core 156 may protrude into the recess 164. Assuming the polishing pad 104 is a two-layer pad, the recess 164 can be constructed by removing a portion of the backing layer 114, or by removing a portion of both the backing layer 114 and the polishing layer 112. Alternatively, the polishing pad may lack such a recess; in this case, the magnetic core 156 of the sensor 154 does not protrude above the top of the platform 102. The in-situ monitoring system 146 may include only one sensor 154 (e.g., see...). Figure 1 ).

[0041] Or, such as Figure 2 To show more clearly, the in-situ monitoring system 146 (such as...) Figure 1 The platform 102 (as shown) may include multiple sensors 154 mounted in, on, or under the platform 102, as illustrated in 154a-154f. Although six sensors are shown, it should be understood that in some embodiments, these sensors 154 may include one, two, three, four, five, six, seven, eight, nine, ten, or even more sensors. As will be discussed in more detail below, the number of sensors 154 may be selected based on the size of the platform, the size of the substrate, or a combination thereof, in order to maintain constant monitoring of the substrate 130. That is, as... Figure 2As shown, the number of sensors and their spacing can be selected such that at least one sensor 154 is always positioned below the substrate 130. By utilizing this arrangement, slippage events can be detected and alerted even with a full platform rotation or fewer, such as with five or fewer sensors, such as about four or fewer sensors, such as even about three or fewer sensors sweeping under the carrier head 128. That is, in some aspects, an alarm can be triggered after the first sensor notices a discrepancy in thickness or edge detection, and this alarm can be confirmed by two or even only one additional sensor.

[0042] Sensors 154 may be positioned at equal angular intervals around the central axis 106. Sensors 154 may be equidistantly positioned around the central axis 106, or sensors 154 may be positioned at varying distances from the central axis 106. Providing multiple sensors 154 can increase the rate of data collection. The controller 148 may include demultiplexing functionality in software to select appropriate signals (e.g., selecting each sensor as it travels beneath the substrate 130), or demultiplexing may be provided by hardware components. In some embodiments, each sensor 154 may be positioned in a separate recess 150 for the in-situ monitoring system 146. Alternatively, each of these sensors 154 may be located in the same recess 150, such as when a circular cross-section recess extends beneath the polishing pad 104 through the platform 102.

[0043] Nevertheless, as shown in the figure, the bearing head 128 disposed above the substrate 130 (such as...) Figure 1 As shown, the substrate 130 is pressed against the polishing pad 104 disposed on the platform 102. Figure 1 As discussed above, the carrier head 128 can rotate clockwise or counterclockwise. Alternatively, the platform 102 rotates about the central axis 106. In the illustrated embodiment, the platform 102 rotates counterclockwise. However, it should be understood that the platform 102 can rotate clockwise or counterclockwise because, unlike optical detection methods, sensor orientation does not require adjustment. Furthermore, in some embodiments, the platform 102 and the carrier head 128 can rotate in the same direction or in opposite directions.

[0044] In some embodiments, the platform 102 and sensors 154a-f can be shaped and sized such that the substrate 130 is positioned above at least one sensor 154, such as above at least two of the sensors 154a-f. The positioning of the substrate 130 above one or more sensors 154 can continue during the rotation of the platform 102. For example, as the platform 102 rotates, sensor 154c may move toward the position of sensor 154b, while sensor 154d may move toward the position of sensor 154c, and so on. However, as described above, in some embodiments, at any point during the rotation of the platform 102, only one sensor 154 may be located below the substrate 130. Therefore, as will be discussed in more detail below, in-situ monitoring alarms can be received very quickly due to continuous contact, as two or more sensors can quickly report and verify slippage of the substrate 130.

[0045] refer to Figure 3 In some embodiments, sensor 154 can be operated by an electromagnetic field 320. That is, circuitry 308 applies an AC current to coil 306, which generates a magnetic field 320 between the poles 305a and 305b of magnetic core 304. In operation, a portion of the magnetic field 320 can extend through the polishing pad 104 (see...). Figure 1 The magnetic core 156 may include two, three, four, or more tracks 305 extending parallel to the core backing 318. Alternatively, it may be implemented using only one track (without using the core backing).

[0046] Figure 3 An example of a drive and sensing circuit system 308 is shown. Circuit system 308 includes a capacitor 310 connected in parallel with a coil 306. The coil 306 and capacitor 310 together form an LC resonant slot. In operation, a current generator 312 (e.g., a current generator based on an edge oscillator circuit) drives the system at the resonant frequency of the LC slot circuit formed by the coil 306 (with inductance L) and the capacitor 310 (with capacitance C). The configuration of the coil 306, core 304, and drive and sensing circuit system 308 can have a resonant frequency of approximately 10 kHz to 100 MHz, for example, 10 kHz to 300 kHz. The current generator 312 can be designed to maintain the interpeak amplitude of a sinusoidal oscillation at a constant value. A rectifier 314 rectifies the time-dependent voltage V0 with amplitude and provides it to a feedback circuit 316. The feedback circuit 316 determines the drive current for the current generator 312 to maintain the amplitude of voltage V0 constant.

[0047] As described above, when the platform 102 rotates, one or more sensors 154 sweep across the underside of the carrier head 128 and the substrate 130. By sampling signals from the circuit system 308 at a specific frequency, the circuit system 308 generates measurements at one or more locations on the substrate 130. For each sweep, measurements at one or more locations can be selected or combined. When the magnetic field 320 reaches the substrate 130, the magnetic field 320 can pass through and generate eddy currents in one or more conductive layers of the substrate 130, or can generate current in conductive rings. This creates an effective impedance, thereby increasing the drive current required by the current generator 302 to maintain a constant amplitude of voltage V0. The magnitude of the effective impedance depends on the distance between the sensor 154 and at least one conductive layer of the substrate 130. This distance depends on the thickness of the substrate 130 and the thickness of the polishing pad 104, and more specifically, on the distance between one or more conductive layers of the substrate 130 and their respective sensors 154. Therefore, the drive current generated by the current generator 312 provides a measurement of the thickness of the substrate 130. More specifically, monitoring system 146 can be used to monitor the thickness of one or more conductive layers (e.g., one or more conductive layers of substrate 130) by inducing eddy currents in the conductive layers or generating currents in conductive loops at the conductive layers. Alternatively, monitoring system 146 can be used, for example, by monitoring conductive layers respectively connected to the substrate mounting surface or around the target 328 ( Figure 1 The thickness of the dielectric layer of substrate 130 is monitored by inducing eddy currents or currents in the ring of 122).

[0048] If it is necessary to monitor the thickness of the conductive layer on the substrate, when the magnetic field 320 reaches the conductive layer of the substrate 130, the magnetic field 320 can pass through and generate a current (if a conductive loop is formed in the layer) or generate eddy currents (if the conductive feature is a continuous sheet). This creates an effective impedance, thereby increasing the drive current required by the current generator 312 to maintain a constant amplitude of voltage V0. The magnitude of the effective impedance depends on the thickness of the conductive layer. Therefore, the drive current generated by the current generator 312 provides a measurement of the thickness of the polished conductive layer.

[0049] As described above, if it is necessary to monitor the thickness of the dielectric layer on the substrate, the conductive target 328 can be located on the substrate 130 away from the polished area (e.g., polished between the substrate 130 and the carrier head 128, such as...). Figure 1The target 328 is located on the far side of the dielectric layer (shown in Figure 122). When the magnetic field 320 reaches the conductive target, it can pass through and generate a current (if the target is a ring) or eddy currents (if the target is a sheet). This creates an effective impedance, thereby increasing the drive current required by the current generator 312 to maintain a constant voltage V0 amplitude. The magnitude of the effective impedance depends on the distance between the sensor 154 and the target 328, which in turn depends on the thickness of the dielectric layer milled on a pad of uniform thickness. Therefore, the drive current generated by the current generator 312 provides a measurement of the thickness of the milled dielectric layer.

[0050] Other configurations are also possible for the drive and sensing circuitry system 308. For example, separate drive and sensing coils may be wound around the core, the drive coil may be driven at a constant frequency, and the amplitude or phase of the current from the sensing coil (relative to the drive oscillator) may be used to provide a signal that provides a measurement of the thickness of the substrate 130.

[0051] A controller 148, such as a general-purpose programmable digital computer, receives a sequence of values ​​from an in-situ monitoring system 146. As the sensor 154 scans beneath the substrate 130 with each rotation of the platform 102, information about the depth of the substrate 130, or even individual substrate layers (e.g., one or more conductive layers disposed in a horizontal layer along the substrate 130), is accumulated in-situ (once per platform rotation). The controller 148 can be programmed to sample measurements from the monitoring system 146 when the substrate 130 typically covers the recessed portions 164. As polishing progresses, the thickness of the substrate 130 and / or its individual layers changes, and the sampled signals vary over time. During polishing, measurements from the monitoring system can be displayed on an output device to allow the operator of the device to visually monitor the progress of the polishing operation.

[0052] Figures 4A to 4C The process of grinding the conductive layer is shown. Furthermore, Figure 5 An exemplary graph is shown for the signal 520 from the in-situ monitoring system 146. Signal 520 in... Figure 5 The image is presented in an idealized form; the original signal will include significant noise.

[0053] Initially, such as Figure 4AAs shown, for the polishing operation, substrate 430 is placed in contact with polishing pad 400. Substrate 430 may include silicon wafer 402 and conductive layer 404, which is, for example, a metal disposed on one or more patterned underlayers 406, such as copper, aluminum, cobalt, titanium, or titanium nitride. These patterned underlayers may be semiconductor layers, conductor layers, or insulating layers. Barrier layer 408, such as tantalum or tantalum nitride, may separate the metal layer from the underlying dielectric. Patterned underlayer 406 may include metallic features such as trenches, vias, pads, and interconnects of copper, aluminum, or tungsten. Furthermore, as described above, and although not shown, each substrate 430 may include a plurality of conductive layers 404 separated by patterned underlayer 406 and / or barrier layer 408.

[0054] Because the bulk of conductive layer 404 is initially relatively thick and continuous before grinding, it has low resistance and can generate relatively strong eddy currents within the conductive layer. These eddy currents cause the metal layer to act as an impedance source in parallel with capacitor 310. For example, a signal can begin at time T1 with an initial value V1 (see...). Figure 5 In an implementation, the initial time can be zero point or the start of slip detection. Therefore, in an implementation, this method can instead represent a first signal strength value at a first time, wherein the first signal strength value can be measured at any time when it is desired to begin detecting wafer slip.

[0055] refer to Figure 4B When the substrate 430 is polished, the main portion of the conductive layer 404 is thinned. As the conductive layer 404 is thinned, its sheet resistance increases, and eddy currents in the metal layer are suppressed. Therefore, the coupling between the conductive layer 404 and the sensor circuit system is reduced (i.e., the resistance of the virtual impedance source is increased). In some implementations of the sensor circuit system 308, this can cause the signal to decrease from its initial value V1.

[0056] refer to Figure 4C Finally, the main portion of the conductive layer 404 is removed, leaving conductive interconnects 410 in the trenches between the patterned insulating layers 406. At this point, the coupling between the conductive portions in the substrate (which are typically small and discontinuous) and the signal from the sensor circuitry tends to plateau (although it may continue to decrease with decreasing trench depth). This results in a significantly reduced rate of change of the amplitude of the output signal from the sensor circuitry. Figure 5 As shown, this occurs at time T2 when the signal reaches value V2.

[0057] Back Figure 1 If the goal is to stop grinding when the lower layer is exposed, then the value V2 (see...) Figure 5A threshold can be used for endpoint detection. However, as mentioned above, the signal from the in-situ monitoring system 146 may include noise. Therefore, a filter can be applied to the raw signal from the in-situ monitoring system 146. For example, the controller 148 can apply a filter, such as a notch filter or a moving average filter, to the signal received from the in-situ monitoring system 146 to generate a filtered signal. Other types of filters can be applied, such as bandpass filters, low-pass filters, high-pass filters, integrated filters, or median filters. The filtered signal can then be used for endpoint determination.

[0058] Figure 6 An illustrative graph showing the signal used by the in-situ monitoring system 146. (Reference) Figure 1 and Figure 6 Sensor 154 can generate a "raw" signal 620. Although in Figure 6 The signal 620 is shown as a continuous line, but in reality, the original signal 620 is a discrete sequence of values. Measurements can be acquired at a set frequency. For example, if sensor 154 passes under substrate 130 once per revolution of platform 102, the measurement frequency can be equal to the platform's rotation speed. However, each graph can be used for a single sensor 154, each sensor 154a-f can generate signal 620, or signals from multiple sensors 154a-154f can be normalized for a single output signal 620.

[0059] like Figure 6 As shown, this signal 620 may include significant noise, therefore the controller 148 applies filtering to the signal 620 to produce a filtered signal 640. Again, although shown as a continuous line, the filtered signal 640 can actually be a discrete sequence of values, each value in the sequence being calculated based on a combination of multiple values ​​from the original signal. In some implementations, the filtered signal 640 is generated by fitting a function (e.g., a polynomial function, such as a first- or second-order polynomial function) to the sequence of values.

[0060] As described above, the generation of the filtered value is delayed because the signal value needs to be acquired after the nominal measurement time. For example, assuming the asymmetry of substrate 130 is small and measurements are performed at a regular frequency, if the filter operates by generating an output value that is a moving average of five consecutive values ​​from the original signal, then a given output value will more accurately represent the measurement obtained when the third value is obtained from the original signal rather than when the fifth value is obtained. This is in... Figure 6 The filter signal 640 is shifted to the right relative to the dashed line 635 (this represents the hypothetical filter signal generated without time offset caused by delay).

[0061] To compensate for the time required for the filter to acquire data, the nominal threshold can be adjusted. Specifically, the controller 148 can store a time delay value ΔT representing the time offset generated by the filter. The controller 148 can also determine the slope R of the filtered signal 640. This slope R can represent the current grinding rate. When VT is the original threshold (e.g., from...), Figure 5 In the case of V2), the adjusted threshold VT' can be calculated as

[0062] VT' = VT - (ΔT * R)

[0063] Subsequently, when the filtered signal 640 crosses the adjusted threshold VT', the controller can trigger the endpoint at time TE.

[0064] Or, such as Figure 7 As shown, it is possible to project the filtered signal 640 forward by an amount of time equal to the time delay value ΔT to generate a projected signal 645. When the controller detects that the projected signal 645 crosses the threshold VT at time TE+ΔT, the controller can trigger the endpoint at time TE. This is essentially equivalent to adjusting the threshold.

[0065] In some implementations, the time delay value ΔT can be input by the user. In some implementations, the time delay value ΔT can be automatically calculated by the controller 148 based on the properties of the filter. For example, for an unweighted moving average, the time delay value ΔT can be half the time it takes for the original value to be averaged.

[0066] For a weighted moving average, the time delay value ΔT can be similarly based on the weights. For example, the filtered value x̅ can be calculated as:

[0067]

[0068] Where N is the number of consecutive values ​​being averaged, and a k The weight is assigned to each signal value from the column. In this case, the time delay value ΔT can be calculated as:

[0069]

[0070] Where f is the sampling rate (e.g., the frequency at which the original value is generated, e.g., once per platform rotation).

[0071] Typically, the time delay value can be determined based on the measurement frequency and order of the filters, using techniques applicable to individual filters.

[0072] In some implementations, the user can input the time period during which the filter will operate to the controller 148; in this case, the controller 148 can calculate the time delay value ΔT based on this time period (e.g., half the time period of the unweighted moving average) and can calculate the number of values ​​to be used in the filter based on the sampling rate. In some implementations, the user can input the number of values ​​to be used in the filter to the controller; in this case, the controller 148 can calculate the time delay value ΔT based on the number of values ​​and the sampling rate.

[0073] The above techniques can be applied to values ​​that have been converted to thickness measurements, or to values ​​that have not been converted. For example, controller 148 may include functions, such as polynomial functions or lookup tables, that output thickness values ​​as a function of the measured values ​​(e.g., voltage values ​​or, possibly, a percentage of signal strength). Therefore, Figure 6 and Figure 7 The signal 640 shown can be a sequence of thickness values ​​generated by converting the measured values ​​into thickness values ​​using this function, or a sequence of measured values ​​that depends on the thickness but is not converted into actual thickness values.

[0074] In some implementations, the slope R is calculated in units of measurement, and then the slope R is converted into a grinding rate in units of thickness. For example, if the polynomial function relating thickness Y to the measurement value X is...

[0075]

[0076] Since R = dX / dt, the grinding rate dY / dt can be calculated as follows:

[0077]

[0078] Alternatively, in some implementations, the filtered signal 640 can be converted from a self-measured value to a thickness measurement value to determine the grinding rate (i.e., a function is fitted to the thickness value, but in units of the measured value).

[0079] The controller 148 can also be connected to a pressure mechanism that controls the pressure applied by the bearing head 128, to the bearing head rotation motor 136 to control the bearing head rotation rate, to the platform rotation motor 108 to control the platform rotation rate, or to the slurry distribution system 120 to control the slurry composition supplied to the grinding pad. Specifically, after classifying the measured values ​​into radial ranges, information about the layer thickness can be fed into the dead-loop controller in real time to periodically or continuously modify the grinding pressure distribution applied by the bearing head.

[0080] Nevertheless, it should be clear that, to date, the charts showing the adjusted thickness changing over time (e.g., Figure 6 and Figure 7The diagram shows the measurement results when the substrate 130 is properly positioned between the carrier head 128 and the polishing pad 104. Surprisingly, this disclosure has found that the monitoring system 146 can also be used to detect substrate slippage and even provide system alarms or even automatically terminate the polishing process.

[0081] That is, return to reference. Figure 6 and Figure 7 The filtered signal 640 decreases in a roughly linear manner. As previously mentioned, the unfiltered signal 620 does contain small signal variations between consecutive measurements. However, these variations in the unfiltered signal between consecutive measurements typically result in a decrease in signal strength V of approximately 20% or less, such as approximately 17.5% or less, approximately 15% or less, approximately 12.5% ​​or less, approximately 10% or less, approximately 7.5% or less, or approximately 5% or less.

[0082] Conversely, substrate slippage can even cause a signal strength decrease of approximately 30% or more between consecutive measurements, such as approximately 40% or more, approximately 50% or more, approximately 60% or more, approximately 70% or more, approximately 80% or more, approximately 90% or more, up to 100% of the signal strength. Furthermore, substrate slippage can be detected by a single sensor 154a. However, in some embodiments, such as when the automatic stop of the polishing process is initiated by a decrease in signal strength, the change in signal strength can be confirmed by a second sensor 154b, a third sensor 154c, a fourth sensor 154d, a fifth sensor 154e, a sixth sensor 154f, or a combination thereof, before the automatic termination of the polishing process.

[0083] Unlike conventional optical sensors used in the art, the signal strength V measurement performed by sensor 154 can be applied to opaque, reflective, and non-reflective surfaces. For example, since the signal strength is based on the thickness of substrate 130 or the thickness of one or more of its conductive layers (or the thickness between targets 328), the slip detection method described herein can be used even when the surface of substrate 130 is covered by abrasive liquid or otherwise. Furthermore, because the slip detection method is measured in-situ and can be integrated into the polishing module controller 148, polishing can be stopped once a significant signal drop is observed, or verified by one or more sensors 154 (e.g., automatically by the controller 148). This allows the polishing process to stop almost immediately after slip is detected, preventing damage to substrate 130 and carrier head 128.

[0084] Furthermore, the controller 148 is programmable to divide measurements from the monitoring system 146 during each sweep from below the substrate 130 into multiple sampling regions, calculate the radial position of each sampling region, and classify the measurements into radial ranges. However, "chattering"—inconsistent determination of the measurement position from one scan to another—causes the leading and trailing edges of the trajectory to shift back and forth in the time domain.

[0085] Therefore, in some implementations, as an alternative to thickness monitoring via signal strength, a “pin location” calibrated by moving a substrate without head sweep can be used to determine whether the substrate is correctly positioned below the carrier 128. The pin location can be detected from the first derivative of the substrate edge signal, which is independent of the film profile. Once this pin location is obtained, it is used to calculate the real-time head sweep and sense the substrate position. This technique significantly reduces jitter and allows for more accurate determination of the sensor's position on the substrate. The sensor position can be calculated using sensor measurements from the grinder, rather than relying on process parameter information (e.g., platform rotation rate) sent from the grinder.

[0086] For example, return to reference Figure 1 and Figure 8 to Figure 11 The system can monitor signals from the in-situ monitoring system 146 to detect the leading and trailing edges of the substrate. Alternatively, signals from the in-situ monitoring system 146 can be monitored to detect the leading and trailing edges of the support ring 144, for example, the leading and trailing edges of the inner surface 144a of the support ring 144, or the leading and trailing edges of the outer surface 144b of the support ring 144.

[0087] To detect the leading and trailing edges, the first derivative of the signal can be calculated and monitored. For example, the first derivative of the signal can be calculated and monitored for peaks (for the leading edge of the substrate or the outer surface of the support ring) and valleys (for the trailing edge of the substrate or the outer surface of the support ring). As another example, the first derivative of the signal can be calculated and monitored for valleys (for the leading edge of the inner surface of the support ring) and peaks (for the trailing edge of the inner surface of the support ring). The times of occurrence of the peaks and valleys indicate the time it takes for the sensor to pass through the leading and trailing edges, respectively.

[0088] To calculate the measured radial position, the grinding system can initially be operated in calibration mode, in which the bearing head 128 does not oscillate laterally. (Reference) Figure 9 During this calibration run, the bearing head is positioned such that the center of the bearing head 128 is at the same radial distance from the sensor to the axis of rotation of the platform 102.

[0089] As discussed above, based on the signal received by the self-monitoring system 146, the controller 148 detects the time t when the sensor crosses the leading edge. LE ( Figure 9 t in 前缘 Similarly, the time t for detecting the sensor crossing the trailing edge is also detected. TE ( Figure 9 t 后缘 The platform rotation rate ω can be calculated based on signals from position sensors. Alternatively or additionally, ω is taken from control values ​​stored in the controller.

[0090] Based on this equivalence and the law of cosines, the radial position r of the "pin point" can be calculated using the following formula. 针尖 :

[0091]

[0092] Where HS is the head sweep position (the distance between the rotation axis of platform 102 and the central axis 138 of the bearing head), and r 传感器 This is the known distance between the sensor and the axis of rotation of the platform. Here, the term "pin point" refers to a set point on an edge, such as the edge of the substrate or the inner or outer surface of the support ring.

[0093] In subsequent monitoring steps, the measurement position can be calculated based on the location of the pin point. If the edge of the support ring is used as the pin point, there may be no substrate during calibration. HS and r during calibration operation 传感器 The indicative value for both is 7.5 inches.

[0094] refer to Figure 11 For the polishing of the substrate, the polishing system can initially operate in normal mode, in which the bearing head 128 oscillates laterally and the substrate 130 is monitored by the in-situ monitoring system 146. In this mode, the head sweep position HS can be calculated on a sweep-by-sweep basis. That is, for each sweep, time t is determined based on the signal from the eddy current monitoring system. LE and t TE The head sweep position HS can be determined based on ω and t. LE t TE 、r 针尖 and r 传感器 Calculate using the equations above and below, where a = 1:

[0095]

[0096] Then, based on HS, ω, t LE t TE and r 传感器 And the specific time t at which the measurement occurs (in real time) 快闪 Based on individual measurements, the position of each measurement by the in-situ monitoring system, i.e., the radial distance d from the measurement to the center of the substrate, is calculated using the following equation:

[0097]

[0098] γ represents the angle between the sensor and the line connecting the center of the platform 102 and the center 138 of the carrier head 128 during measurement. Similarly, the platform rotation rate ω can be calculated based on the signal from the position sensor. Alternatively or additionally, ω can be taken from a control value stored in the controller.

[0099] By using geometric calculations of the pinpoint location and sensor position on the substrate, the actual measurement location (e.g., radial position relative to the substrate center) can be determined more accurately, thus reducing jitter. This enables improved scan-to-scan and sensor-to-sensor matching. Consequently, endpoint determination becomes more reliable and / or wafer uniformity can be improved.

[0100] Furthermore, as described above, it was surprisingly found that pinpoint calculation can be used to determine substrate slip events. That is, as can be clearly seen from the above description, pinpoint calculation requires detection of the leading and trailing edges of substrate 130. Therefore, in some aspects, methods for automatically terminating the polishing process can be based entirely or partially on substrate chatter correction or substrate centering correction failures, since no substrate edge is located for pinpoint calculation or similar calculations. For example, with Figure 10 If the center position is different, this failure will result in no centering and / or no edge detection. Unlike conventional optical sensors used in this field, substrate vibration determination can be used with opaque, reflective, and non-reflective surfaces.

[0101] For example, since the jitter value is based on the conductive material at the leading and trailing edges of the substrate, the slip detection method described herein can be used even if the surface of substrate 130 is covered by polishing liquid or otherwise. Furthermore, since the slip detection method is measured in-situ and can be integrated into the polishing module controller 148, polishing can be stopped if one or more edges cannot be positioned, the substrate cannot be centered, or a combination thereof, or when such parameters are verified by one or more sensors 154 (e.g., automatically by the controller 148). This allows the polishing process to stop almost immediately after slip is detected, preventing damage to substrate 130 and carrier head 128. Additionally, it should be clarified that the method discussed herein utilizes measurements of the leading and trailing edges of substrate 130, because determining the leading and trailing edges based on the carrier ring 144 of carrier head 128 would not detect substrate slip events.

[0102] Nevertheless, as stated above, in some embodiments, both the signal strength method and the edge detection method may be used together to provide additional redundancy in slip detection. However, it should be understood that in some embodiments, either the signal strength method or the edge detection method may be sufficient to detect substrate slip.

[0103] The in-situ monitoring system 146 can be used in various polishing systems. The polishing pad, the support head, or both can be movable to provide relative movement between the polishing surface and the substrate. The polishing pad can be a circular (or some other shape) pad fixed to a platform, a tape extending between the supply roller and the take-up roller, or a continuous belt. The polishing pad can be fixed to the platform and gradually advance above the platform between polishing operations, or continuously driven above the platform during polishing. During polishing, the pad can be fixed to the platform, or there may be a fluid bearing between the platform and the polishing pad during polishing. The polishing pad can be a standard (e.g., polyurethane with or without filler) rough pad, soft pad, or fixed polishing pad.

[0104] In the foregoing description, numerous specific details have been set forth for illustrative purposes and to provide a comprehensive understanding of the various embodiments. However, it will be apparent that some embodiments may be practiced without some of these specific details. In other instances, well-known structures and apparatuses are shown in block diagram form.

[0105] The foregoing description provides only illustrative embodiments and is not intended to limit the scope, applicability, or configuration of this disclosure. Rather, the foregoing description of various embodiments provides an enabling disclosure for implementing at least one embodiment. It should be understood that various changes may be made to the function and arrangement of the elements without departing from the spirit and scope of some of the embodiments set forth in the appended claims.

[0106] Specific details are set forth in the foregoing description to provide a comprehensive understanding of the implementation. However, it should be understood that implementations may be practiced without such specific details. For example, circuits, systems, networks, processes, and other components may be shown as components in block diagram form to avoid obscuring the implementation with unnecessary details. In other instances, well-known circuits, processes, algorithms, structures, and techniques may be shown without unnecessary details to avoid obscuring the implementation.

[0107] Furthermore, it should be noted that individual implementations may be described as processes shown in flow diagrams, flowcharts, data flow diagrams, structure diagrams, or block diagrams. Although flow diagrams may describe operations as sequential processes, many operations can be performed in parallel or simultaneously. Furthermore, the order of operations can be rearranged. A process terminates when its operations are completed, but there may be other steps not included in the diagram. A process may correspond to a method, function, program, subroutine, subprogram, etc. When a process corresponds to a function, its termination may correspond to the function returning to the calling function or the main function.

[0108] As used herein and in the appended claims, the singular forms “a / an” and “the” include plural references unless the context clearly specifies otherwise. Thus, for example, reference to “a heater” includes a plurality of such heaters, and reference to “the protrusion” includes reference to one or more protrusions and their equivalents known to those skilled in the art, and so on.

[0109] As used herein, the terms “about,” “approximately,” or “substantially” may be interpreted as within the range that a person skilled in the art would expect from the description.

[0110] Furthermore, the terms “comprise / comprising,” “contain / containing,” and “include / including” are used in this specification and the following claims to specify the presence of the stated feature, integer, component, or operation, but do not exclude the presence or addition of one or more other features, integers, components, operations, actions, or groups.

[0111] The term "computer-readable medium" includes, but is not limited to, portable or fixed storage devices, optical storage devices, wireless channels, and various other media capable of storing, containing, or carrying instructions and / or data. A code segment or machine-executable instruction may represent a program, function, subroutine, program, routine, subroutine, module, software package, class, or any combination of instructions, data structures, or program descriptions. A code segment can be coupled to another code segment or hardware circuit by transmitting and / or receiving information, data, arguments, parameters, or memory contents. Information, arguments, parameters, data, etc., can be transmitted, forwarded, or transported in any suitable manner, including memory sharing, message passing, symbol passing, network transmission, etc.

[0112] Furthermore, the implementation can be carried out by hardware, software, firmware, middleware, microcode, hardware description language, or any combination thereof. When implemented in software, firmware, middleware, or microcode, the program code or code segments used to perform the necessary tasks can be stored in a machine-readable medium. The processor can then perform the necessary tasks.

[0113] In the foregoing description, features have been described with reference to specific embodiments thereof, but it should be understood that not all embodiments are limited thereto. Various features and aspects of some embodiments may be used alone or in combination. Furthermore, embodiments may be used in any number of environments and applications beyond those described herein without departing from the broader spirit and scope of this specification. Therefore, the description and drawings should be considered illustrative rather than restrictive.

[0114] Furthermore, for ease of explanation, the methods have been described in a specific order. It should be understood that, in alternative embodiments, these methods may be performed in a different order than described. It should also be understood that the above methods may be performed by hardware components or may be embodied in a sequence of machine-executable instructions that can be used to cause a machine (such as a general-purpose or special-purpose processor or instruction-programmed logic circuit) to perform these methods. These machine-executable instructions may be stored on one or more machine-readable media, such as CD-ROMs or other types of optical discs, floppy disks, ROM, RAM, EPROM, EEPROM, magnetic cards or optical cards, flash memory, or other types of machine-readable media suitable for storing electronic instructions. Alternatively, these methods may be performed by a combination of hardware and software.

Claims

1. A method for detecting substrate slippage, the method comprising: Make the surface of one layer of the substrate contact the polishing pad; This causes relative movement between the substrate and the polishing pad; When the layer of the substrate undergoes grinding using a rotatable platform, the first and second sensors of a plurality of sensors in the in-situ monitoring system are swept across the substrate. The in-situ monitoring system generates a first signal value sequence from the first sensor and a second signal value sequence from the second sensor, wherein the first signal value sequence and the second signal value sequence include signal strength relative to the thickness of the layer; The time it takes for the first sensor to cross the leading edge of the substrate and the time it takes for the first sensor to cross the trailing edge of the substrate are detected from the first signal value sequence of the first sensor from the in-situ monitoring system. The second time when the second sensor crosses the leading edge of the substrate and the second time when the second sensor crosses the trailing edge of the substrate are detected from the second signal value sequence of the second sensor from the in-situ monitoring system. For each signal value in at least some of the first signal value sequences, the position of the signal value on the substrate is determined based on the time it takes for the first sensor to cross the leading edge of the substrate and the time it takes for the first sensor to cross the trailing edge of the substrate. For each of at least some of the second signal values ​​in the second signal value sequence, the position of the second signal value on the substrate is determined based on the time it takes for the second sensor to cross the leading edge of the substrate and the time it takes for the second sensor to cross the trailing edge of the substrate. and If the signal strength changes by 30% or more from the first signal value sequence to the second signal value sequence, and the position of the second signal value on the substrate, or a combination thereof, cannot be determined, then a slip alarm is activated.

2. The method of claim 1, wherein determining the position comprises: determining the first derivative of the first signal value sequence; and identifying valleys and peaks in the first derivative.

3. The method of claim 1, further comprising a third signal value sequence from a third sensor among the plurality of sensors, wherein the third sensor is disposed between the first sensor and the second sensor.

4. The method of claim 3, wherein the slip alarm is activated only if the signal strength changes by 30% or more from the first signal value sequence to the third signal value sequence and from the third signal value sequence to the second signal value sequence, and the position of the second signal value and the third signal value on the substrate, or a combination thereof, cannot be determined.

5. The method of claim 3, wherein the slip alarm stops the relative motion upon activation.

6. The method of claim 5, wherein the relative motion is stopped by a controller.

7. The method of claim 1, wherein detecting the signal value comprises: detecting conductive layers within the leading and trailing edges of the substrate.

8. A method for detecting substrate slippage, the method comprising: Make the surface of one layer of the substrate contact the polishing pad; This causes relative movement between the substrate and the polishing pad; The layer of the substrate is monitored using an in-situ monitoring system to generate a signal, wherein the signal strength depends on the thickness of the layer; The signal is filtered to generate a filtered signal; and If the second signal strength value differs from the first signal strength value by 30% or more, a slip alarm will be triggered.

9. The method of claim 8, wherein the in-situ monitoring system comprises a plurality of sensors, wherein the plurality of sensors generate the signal.

10. The method of claim 9, wherein at least one of the plurality of sensors, a first sensor, is disposed in a platform located below the abrasive pad.

11. The method of claim 9, wherein the plurality of sensors are positioned to intermittently scan under the substrate such that at least one sensor is positioned under the substrate at any given time.

12. The method of claim 11, wherein the relative motion is stopped by a controller.

13. The method of claim 12, wherein the relative motion is stopped if the slip alarm is activated.

14. The method of claim 11, wherein the plurality of sensors includes at least one first sensor, a second sensor, and a third sensor.

15. The method of claim 14, wherein the slip alarm is activated only if the signal strength changes by 30% or more from both the first signal strength value to the second signal strength value and the second signal strength value to the third signal strength value.

16. A computer program, tangibly encoded on a non-transitory computer-readable medium, the computer program comprising instructions that cause a computer system to perform the following: Receive a first signal value sequence from a first sensor of an in-situ monitoring system and a second signal value sequence from a second sensor of the in-situ monitoring system, wherein the first signal value sequence and the second signal value sequence sequentially include signal strength relative to the thickness of a layer of the substrate, wherein the first sensor and the second sensor scan and monitor the substrate during polishing; The time it takes for the first sensor to cross the leading edge of the substrate and the time it takes for the first sensor to cross the trailing edge of the substrate are detected from the first signal value sequence of the first sensor from the in-situ monitoring system. The second time when the second sensor crosses the leading edge of the substrate and the second time when the second sensor crosses the trailing edge of the substrate are detected from the second signal value sequence of the second sensor from the in-situ monitoring system. For each signal value in at least some of the first signal value sequences, the position of the signal value on the substrate is determined based on the time it takes for the first sensor to cross the leading edge of the substrate and the time it takes for the first sensor to cross the trailing edge of the substrate. For each of at least some of the second signal values ​​in the second signal value sequence, the position of the second signal value on the substrate is determined based on the time it takes for the second sensor to cross the leading edge of the substrate and the time it takes for the second sensor to cross the trailing edge of the substrate. and If the signal strength changes by 30% or more from the first signal value sequence to the second signal value sequence, and the position of the second signal value on the substrate, or a combination thereof, cannot be determined, then a slip alarm is activated.

17. The computer program of claim 16, wherein if the slip alarm is activated, the instruction stops the grinding.

18. The computer program of claim 17, wherein the computer system automatically stops the grinding.

19. The computer program of claim 17, wherein the computer program comprises instructions to cause the computer system to perform the following: The system receives a third signal value sequence from a third sensor of the in-situ monitoring system, the signal values ​​including signal strength relative to the thickness of the layer, wherein the third sensor is circumferentially disposed between the first sensor and the second sensor. The slip alarm is activated only if the signal strength changes by 30% or more from the first signal value sequence to the third signal value sequence and from the third signal value sequence to the second signal value sequence, and the position of the second signal and the third signal value on the substrate, or a combination thereof, cannot be determined.

20. The computer program of claim 16, wherein the instruction to determine the position of the signal value on the substrate includes instructions to determine the distance between the center of the bearing head holding the substrate and the axis of rotation of the rotatable platform based on the time the first sensor crosses the leading edge and the time the first sensor crosses the trailing edge, and instructions to determine the position of the first signal value on the substrate based on the distance.

Images