Substrate polishing apparatus and film thickness calculation method
The apparatus addresses the challenge of measuring film thickness in CMP processes by using an eddy current sensor to correct for noise from metal structures, ensuring accurate and uniform film thickness distribution during polishing.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- EBARA CORP
- Filing Date
- 2024-12-27
- Publication Date
- 2026-07-09
AI Technical Summary
Existing CMP apparatuses face challenges in accurately measuring the thickness of a film to be polished due to the influence of local metal structures on the substrate, which affect the output signal of eddy current sensors.
A substrate polishing apparatus equipped with a rotatable polishing table and head, utilizing an eddy current sensor to acquire waveform data, identify equidistant points, correct the data based on a minimum value, and calculate film thickness while compensating for noise from metal structures using algorithms to ensure accurate measurement.
The solution enables highly accurate film thickness distribution measurement, unaffected by local metal structures, allowing for precise polishing control and uniform film thickness across the substrate.
Smart Images

Figure 2026115240000001_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a substrate polishing apparatus and a film thickness calculation method.
Background Art
[0002] One of the manufacturing apparatuses for semiconductor devices is a CMP (Chemical Mechanical Polishing) apparatus. A typical CMP apparatus includes a polishing table to which a polishing pad is attached and a polishing head to which a substrate is attached. In a typical CMP apparatus, a polishing liquid is supplied to the polishing pad, and at least one of the polishing table and the polishing head is rotated while the polishing pad and the substrate are in contact with each other, thereby polishing the substrate.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] In order to measure the thickness of a film to be polished during polishing of a substrate, an eddy current sensor can be used. The eddy current sensor is provided, for example, on the polishing table. The eddy current sensor moves along a certain orbit on the surface of the substrate as the polishing table rotates, and measures the film thickness at each point on the orbit (see, for example, Patent Document 1). However, if there are local metal structures in addition to the film to be polished (metal film) on the substrate, it becomes difficult to accurately measure the thickness of the film to be polished that should originally be measured due to the influence.
Means for Solving the Problems
[0005] According to one embodiment, a substrate polishing apparatus is provided, comprising: a polishing table equipped with an eddy current sensor, configured to be rotatable; a polishing head facing the polishing table and configured to be rotatable, the polishing head capable of attaching a substrate to a surface facing the polishing table; and a control unit, wherein the control unit is configured to acquire waveform data of the output signal of the eddy current sensor when the eddy current sensor passes through a plurality of trajectories on the substrate, identify a plurality of corresponding points equidistant from the center of the substrate in the plurality of trajectories on the substrate, compare the values of the output signal of the eddy current sensor at the identified plurality of points to find the minimum value, correct the waveform data based on the found minimum value, and calculate the film thickness on the surface of the substrate based on the corrected waveform data. [Brief explanation of the drawing]
[0006] [Figure 1] This is a front view of a substrate polishing apparatus according to one embodiment. [Figure 2] This is a schematic cross-sectional view showing the structure of an exemplary substrate that is to be polished by a substrate polishing device. [Figure 3] This is a flowchart showing the algorithm for a method according to one embodiment of the present invention. [Figure 4] This is a schematic diagram showing an exemplary trajectory of an eddy current sensor passing across a substrate. [Figure 5] This is an example of waveform data. [Figure 6] This is a schematic diagram illustrating the multiple points identified in step 306. [Figure 7] This is an explanatory diagram of the correction process in step 310 for the waveform data of the output signal of the eddy current sensor. [Figure 8] This is a flowchart of the algorithm for a method according to another embodiment of the present invention. [Figure 9] This is an example of waveform data to illustrate the process in step 806. [Modes for carrying out the invention]
[0007] Embodiments of the present invention will be described below with reference to the drawings. In the drawings described below, the same or corresponding components are denoted by the same reference numerals, and redundant descriptions are omitted.
[0008] Figure 1 is a front view of a substrate polishing apparatus 100 according to one embodiment. The substrate polishing apparatus 100 shown in Figure 1 is a CMP (Chemical Mechanical Polishing) apparatus. However, the substrate polishing apparatus 100 is not limited to a CMP apparatus. The substrate polishing apparatus 100 can be any apparatus that polishes a substrate by rotating a polishing table equipped with an eddy current sensor.
[0009] The CMP apparatus 100 comprises a polishing table 110, a polishing head 120, and a liquid supply mechanism 130. The CMP apparatus 100 further comprises a control unit 140 for controlling each component. The control unit 140 may include, for example, a storage device 141, a processor 142, and an input / output device 143.
[0010] A polishing pad 111 is detachably attached to the upper surface of the polishing table 110. Here, the upper surface of the polishing table 110 refers to the surface of the polishing table 110 that faces the polishing head 120. Therefore, the upper surface of the polishing table 110 is not limited to the surface located in the vertically upward direction. The polishing head 120 is positioned to face the polishing table 110. A substrate 121 is detachably attached to the surface of the polishing head 120 that faces the polishing table 110. The liquid supply mechanism 130 is configured to supply a polishing liquid such as slurry to the polishing pad 111. The liquid supply mechanism 130 may also be configured to supply cleaning liquid or chemical liquid in addition to polishing liquid.
[0011] The CMP device 100 can lower the polishing head 120 by a vertical movement mechanism (not shown) to bring the substrate 121 into contact with the polishing pad 111. However, the vertical movement mechanism may also be capable of moving the polishing table 110 up and down. The polishing table 110 and the polishing head 120 are rotated by a motor (not shown) or the like. The CMP device 100 polishes the substrate 121 by rotating both the polishing table 110 and the polishing head 120 while the substrate 121 and the polishing pad 111 are in contact.
[0012] The CMP apparatus 100 may further include an airbag 122 divided into a plurality of concentric circular compartments. The airbag 122 may be located on the polishing head 120. Additionally or alternatively, the airbag 122 may be located on the polishing table 110. The airbag 122 is a component for adjusting the polishing pressure of the substrate 121 for each region of the substrate 121. The airbag 122 is configured so that its volume changes depending on the pressure of the air introduced inside. A fluid other than air, such as nitrogen gas or pure water, may be introduced into the airbag 122.
[0013] An eddy current sensor 150 is installed inside the polishing table 110. The eddy current sensor 150 is positioned so that it passes through the center of the substrate 121 when the polishing table 110 rotates during polishing. The eddy current sensor 150 is configured to induce eddy currents in the conductive layer on the surface of the substrate 121. The eddy current sensor 150 is further configured to receive the change in impedance caused by the magnetic field generated by the eddy currents and output a signal corresponding to the thickness of the conductive layer on the surface of the substrate 121. By using this output signal from the eddy current sensor 150, the thickness of the film to be polished on the surface of the substrate 121 can be determined.
[0014] Here, the output signal of the eddy current sensor 150 is not only affected by the film to be polished exposed on the outermost surface of the substrate 121 (the film formed over the entire outermost surface of the substrate 121). FIG. 2 is a cross-sectional schematic view showing the structure of an exemplary substrate 121 to be polished by the substrate polishing apparatus 100. As shown in FIG. 2, on the upper surface of this exemplary substrate 121, a dielectric film (a film made of, for example, SiO2 or the like) 202 is formed over the entire surface, and further, a metal film (a film made of, for example, Cu or the like) 204 is formed covering the dielectric film 202 thereon. The metal film 204 is the film to be polished located on the outermost surface of the substrate 121. Also, the substrate 121 may have one or more through electrodes 206 for conducting one surface and the other surface thereof. Further, metal wirings 208 may be embedded in the dielectric film 202 on the substrate 121. When the eddy current sensor 150 passes over or near such through electrodes 206 or metal wirings 208, eddy currents are also induced in these metal structures, so the output signal of the eddy current sensor 150 is affected by that, and the value of the signal changes from the output signal of the eddy current sensor 150 when it passes through a region where there are no through electrodes 206 or metal wirings 208 on the substrate 121. That is, the through electrodes 206 and metal wirings 208, which are local metal structures formed on the substrate 121, can cause noise in the output signal of the eddy current sensor 150. Note that the metal structures are not limited to the through electrodes 206 and metal wirings 208 embedded in the dielectric film 202 as described above, and may also include, for example, wirings and vias exposed on the outermost surface of the substrate 121.
[0015] FIG. 3 is a flowchart showing an algorithm of a film thickness calculation method according to an embodiment of the present invention, which can remove or reduce the noise generated in the output signal of the eddy current sensor 150 due to local metal structures of the substrate 121. The processing of this flowchart may be performed by a processor (for example, the processor 142 of the control unit 140).
[0016] First, in step 302, an output signal of the eddy current sensor 150 with respect to the substrate 121 is acquired. Specifically, while rotating both the polishing head 120 to which the substrate 121 to be polished is attached and the polishing table 110 at a predetermined rotational speed, the output signal is acquired from the eddy current sensor 150. The eddy current sensor 150 moves along an arc-shaped orbit corresponding to the ratio of the rotational speed of the polishing table 110 to the rotational speed of the polishing head 120 with respect to the substrate 121 (i.e., as seen from the substrate 121). Each time the polishing table 110 makes one rotation, the eddy current sensor 150 crosses the surface of the substrate 121 along an arc-shaped orbit with a predetermined curvature determined by the rotational speeds of the polishing table 110 and the polishing head 120, and in the next rotation of the polishing table 110, it passes through an orbit corresponding to another arc with the same curvature as the previous rotation. Therefore, signal values at each point on these numerous arc orbits are continuously obtained from the eddy current sensor 150. Hereinafter, a series of signal values obtained from the eddy current sensor 150 when the eddy current sensor 150 passes through a certain orbit on the substrate 121 is referred to as "waveform data" of the output signal of the eddy current sensor 150.
[0017] FIG. 4 is a schematic diagram showing an exemplary orbit of the eddy current sensor 150 passing over the substrate 121. In the figure, each curved arrow in an arc shape represents one orbit of the eddy current sensor 150 and the direction in which the eddy current sensor 150 advances along the orbit. Also, the reference numerals "1" to "10" in the figure indicate the order of the orbits through which the eddy current sensor 150 passes. As the polishing table 110 and the polishing head 120 rotate, the eddy current sensor 150 moves over the surface of the substrate 121 in the order of the orbits "1" to "10" and outputs the signal value at each point on the orbit. In the example of FIG. 4, the angular interval θ between adjacent orbits is set to 36 degrees, and when the polishing table 110 makes 10 rotations, the eddy current sensor 150 follows the same orbit as before (after the orbit "10", it becomes the orbit "1" again). Note that the angular interval θ of the orbits can be arbitrarily set according to the combination of the rotational speed of the polishing table 110 and the rotational speed of the polishing head 120.
[0018] Figure 5 shows an example of waveform data obtained in step 302. The horizontal axis of the graph in Figure 5 represents eddy electricity. The graph shows the position of the flow sensor 150 on its trajectory (i.e., the distance from the center of the substrate 121 to the eddy current sensor 150), and the vertical axis shows the signal value of the eddy current sensor 150. The waveform data corresponds to one trajectory. As the polishing table 110 and polishing head 120 rotate, waveform data like that shown in Figure 5 is sequentially acquired each time the eddy current sensor 150 passes through one trajectory.
[0019] Next, in step 304, the control unit 140 determines whether waveform data has been acquired for a predetermined number of tracks. The "determined number" may be the number of tracks that can evenly cover almost the entire surface of the substrate 121. For example, in the example of tracks shown in Figure 4 above, if the number of tracks is 5 or more (for example, 5 tracks from track "1" to "5"), it is possible to evenly cover almost the entire surface of the substrate 121. In this case, in step 304, it may be determined whether waveform data has been acquired for 5 tracks (or more tracks). The number of tracks that can evenly cover almost the entire surface of the substrate 121 may vary depending on the rotation speed of the polishing table 110 and the polishing head 120. For example, in the example of Figure 4 above, the angular spacing θ between adjacent tracks is set to 36 degrees, but if the difference between the rotation speed of the polishing table 110 and the rotation speed of the polishing head 120 becomes small, the angular spacing θ between the tracks becomes narrower than 36 degrees, and more tracks will be needed to evenly cover the entire surface of the substrate 121. Furthermore, if the angular spacing θ of the trajectories is narrow, the number of measurement points on the substrate 121 increases, thereby improving the spatial resolution of the film thickness measurement. However, the time it takes for the trajectories to evenly cover the entire surface of the substrate 121 also increases. On the other hand, if the angular spacing θ of the trajectories is set wider, the spatial resolution of the film thickness measurement decreases, but the trajectories can cover the entire surface of the substrate 121 in a shorter time, allowing for a quicker understanding of the film thickness across the entire surface of the substrate 121.
[0020] Once waveform data for a predetermined number of trajectories has been acquired, the process proceeds to step 306. In step 306, the control unit 140 identifies a plurality of corresponding points on the plurality of trajectories on the substrate 121 that are equidistant from the center of the substrate 121. For example, Figure 6 shows three trajectories on the substrate 121 through which the eddy current sensor 150 passes, namely trajectory "1", trajectory "2", and trajectory "3". In this example, point P1 on trajectory "1", point P2 on trajectory "2", and point P3 on trajectory "3" are equidistant from the center of the substrate 121 by radius r i These are corresponding points that are equidistant from each other. The control unit 140 can control any radius r from the center of the substrate 121. i Regarding this, we identify several points that correspond to each other. It goes without saying that Figure 6 is for illustrative purposes only, and in reality, there may be more than three orbitals on the substrate 121.
[0021] Next, in step 308, the control unit 140 compares the output signal values of the eddy current sensor 150 at a plurality of corresponding points identified in step 306 to find the minimum value. Specifically, referring again to Figure 6, the control unit 140 compares the output signal value of the eddy current sensor 150 at point P1 on orbit "1", the output signal value of the eddy current sensor 150 at point P2 on orbit "2", and the output signal value of the eddy current sensor 150 at point P3 on orbit "3", and determines the minimum value among them. The control unit 140 does the same for any radius r i We will implement the following.
[0022] Next, in step 310, the control unit 140 corrects the waveform data based on the minimum value found in step 308. Correcting the waveform data involves applying the minimum value found in step 308 to each point on the waveform data. As described above, in step 308, any radius r from the center of the substrate 121 i Regarding the radius r, iThe minimum value is determined from the output signal values of the eddy current sensor 150 at multiple points corresponding to each radius. In other words, one minimum signal value of the eddy current sensor 150 is determined for each of the different radii. In step 310, the waveform data is corrected by replacing the signal value at each point on the waveform data (i.e., the signal value of the eddy current sensor 150 obtained at each radial position on the substrate 121) with the minimum signal value corresponding to that point (i.e., the minimum value found in step 308). It will take place.
[0023] Figure 7 shows examples of waveform data before and after correction to illustrate the correction process in step 310. In Figure 7, the waveform data D1 to D10 before correction are shown as solid lines. The waveform data D1 to D10 before correction correspond to, for example, trajectories "1" to "10" in Figure 4. In Figure 7, the waveform data after correction D correct This is indicated by a thick dotted line obtained by tracing only the smallest point of the uncorrected waveform data D1 to D10.
[0024] The numerous peaks in the uncorrected waveform data are likely noise originating from locally present metal structures (through electrodes 206 and metal wiring 208) on or inside the substrate 121. Therefore, by correcting the waveform data according to steps 306-310, noise generated in the output signal of the eddy current sensor 150 due to localized metal structures in the substrate 121 can be removed or reduced.
[0025] Next, in step 312, the control unit 140 calculates a moving average of the waveform data corrected in step 310 with respect to the radial direction of the substrate 121 (i.e., a moving average with respect to the horizontal axis direction in the graph of Figure 7). The processing in steps 306 to 312 is repeated each time the eddy current sensor 150 passes through a new trajectory. For example, steps 306 to 312 are performed when the eddy current sensor 150 has passed through 10 trajectories "1" to "10". When the eddy current sensor 150 passes through trajectory "1" again after trajectory "10", steps 306 to 312 are similarly performed using the waveform data corresponding to the latest 10 trajectories (i.e., 10 trajectories including trajectories "2" to "10" and trajectory "1" which was passed after trajectory "10"). The same applies thereafter.
[0026] Next, in step 314, the control unit 140 calculates a moving average of the multiple waveform data (after radial moving average) sequentially calculated in step 312 each time the eddy current sensor 150 passes through a new trajectory. For example, the moving average may be calculated for multiple waveform data including the waveform data obtained by performing the processing in steps 306 to 312 using the first waveform data D1 to D10, the waveform data obtained by performing the processing in steps 306 to 312 using the next waveform data D2 to D11, and the waveform data obtained by performing the processing in steps 306 to 312 using the next waveform data D3 to D12. The moving average processing in steps 312 and 314 can remove fine noise present in the waveform data of the output signal of the eddy current sensor 150.
[0027] Next, in step 316, the control unit 140 calculates the film thickness on the surface of the substrate 121 based on the waveform data after processing in step 314.
[0028] As a result, highly accurate radial film thickness distribution data for the substrate 121 can be obtained, which is not affected by the through-electrodes 206 or metal wiring 208 of the substrate 121. From the film thickness distribution data obtained during polishing of the substrate 121, the control unit 140 can accurately determine the polishing endpoint. Alternatively, the control unit 140 may increase or decrease the internal pressure of the airbag 122 based on the film thickness distribution data obtained during polishing of the substrate 121, increasing the polishing pressure in areas with thicker film thickness (i.e., areas with low polishing progress) and decreasing the polishing pressure in areas with thinner film thickness (i.e., areas with high polishing progress). This control makes it possible to make the film thickness of the substrate 121 uniform.
[0029] Figure 8 is a flowchart showing an algorithm for a film thickness calculation method according to another embodiment of the present invention, which can remove or reduce noise generated in the output signal of the eddy current sensor 150 due to localized metallic structures on the substrate 121. The processing in this flowchart is a process This may be carried out by a control unit (for example, the processor 142 of the control unit 140). Steps 302 and 304 of the algorithm according to this embodiment are the same as those of the embodiment in Figure 3 described above, and redundant explanations will be omitted. However, in step 304 of this embodiment, the "predetermined number" only needs to be "at least three".
[0030] In step 806, following step 304, the control unit 140 determines whether or not there is a metal structure on the track of the substrate 121 based on a comparison of the similarity of the multiple waveform data acquired in step 304. The control unit 140 may determine that there is a metal structure on the track corresponding to the remaining waveform data if a predetermined number (e.g., a majority) of the multiple waveform data have a high degree of similarity and the remaining waveform data have a low degree of similarity.
[0031] For example, suppose in step 304, three waveform data sets D1, D2, and D3 are acquired as shown in Figure 9. The waveform data sets D1, D2, and D3 may correspond to, for example, trajectories "1," "2," and "3," respectively. In this example, waveform data sets D1 and D2 have a relatively flat shape over almost the entire trajectory. In contrast, waveform data set D3 has two peaks 902 and 904 in the middle of the trajectory. These peaks are likely due to noise from a metal structure (through electrode 206 or metal wiring 208) locally present on or inside the substrate 121. Therefore, by comparing the shapes of the waveform data sets D1, D2, and D3, it can be determined that a metal structure exists at the position corresponding to the peak on trajectory "3" among the multiple trajectories that the eddy current sensor 150 has passed through, based on the fact that waveform data set D3 has a shape with a different tendency than waveform data sets D1 and D2. In this way, it is possible to find the trajectory in which a metal structure exists by comparing the similarity of multiple waveform data sets and using "majority voting." Furthermore, if all the waveform data have similar shapes, it can be determined that no trajectories passed over metal structures.
[0032] The comparison of the similarity of multiple waveform data in step 806 can be performed using various methods. For example, the correlation coefficient of multiple waveform data D1, D2, D3, ... can be calculated, and waveform data with a low correlation coefficient can be determined to have low similarity. Alternatively, machine learning using clustering methods such as k-means or One Class SVM (Support Vector Machine) can be used to detect peak portions (e.g., peaks 902 and 904 in Figure 9) in multiple waveform data D1, D2, D3, .... Furthermore, an average curve of multiple waveform data can be calculated by calculating the average value between multiple waveform data for each point on the waveform data, and waveform data with different trends can be identified from the multiple waveform data based on the difference between the average curve and each waveform data.
[0033] If it is determined in step 806 that a metal structure is in the trajectory of the eddy current sensor 150, the process proceeds to step 808. In step 808, the control unit 140 corrects the waveform data corresponding to the trajectory in which the metal structure was determined to be present, by reducing the signal value at the location where the metal structure was determined to be present. Specifically, referring again to the example in Figure 9, the waveform data is corrected by reducing the values of two peaks 902 and 904 and their vicinity in the waveform data D3. The correction may be, for example, by changing the signal values of the peaks and their vicinity in the waveform data before correction to a predetermined value that is smaller than the peak value of each peak. The predetermined value may be, for example, the peak value of each peak multiplied by a predetermined attenuation rate (e.g., 30%).
[0034] As mentioned above, the peaks in the waveform data for the embodiment shown in Figure 3 are highly likely to be noise from metal structures (through electrodes 206 and metal wiring 208) locally present on or inside the substrate 121. Therefore, by correcting the waveform data in steps 806 and 808, eddy currents caused by localized metal structures in the substrate 121 can be eliminated. Noise occurring in the output signal of the sensor 150 can be removed or reduced.
[0035] Next, in step 810, the control unit 140 calculates the film thickness on the surface of the substrate 121 based on the waveform data corrected in step 808 and the waveform data other than the corrected waveform data. For example, in the example of Figure 9 above, average waveform data may be created by averaging the waveform data D3 corresponding to the trajectory where the metal structure exists, which has been corrected in the process of step 808, with the (uncorrected original) waveform data D1 and D2 corresponding to the trajectory where the metal structure does not exist, and the film thickness on the surface of the substrate 121 may be calculated using this average waveform data.
[0036] While embodiments of the present invention have been described above based on several examples, the embodiments described above are intended to facilitate understanding of the present invention and do not limit it. The present invention can be modified and improved without departing from its spirit, and of course, its equivalents are included. Furthermore, any combination or omission of the components described in the claims and specification is possible to the extent that at least some of the above-mentioned problems can be solved or at least some of the effects can be achieved. [Explanation of Symbols]
[0037] 100 Substrate polishing equipment 110 Polishing Table 111 Polishing Pad 120 Polishing Heads 121 circuit boards 122 Airbags 130 Liquid supply mechanism 140 Control Unit 141 Storage Devices 142 processors 143 Input / Output Devices 150 Eddy Current Sensor 202 Dielectric film 204 Metal film 206 Through electrode 208 Metal wiring
Claims
1. A polishing table equipped with an eddy current sensor, comprising a polishing table configured to be rotatable, A polishing head that is rotatably configured to face the polishing table, and to which a substrate can be attached to the surface facing the polishing table, Control unit and A substrate polishing apparatus comprising, The control unit, When the eddy current sensor passes through multiple trajectories on the substrate, waveform data of the output signal of the eddy current sensor is acquired. In the plurality of tracks on the substrate, a plurality of corresponding points that are equidistant from the center of the substrate are identified, The values of the output signals of the eddy current sensors at the aforementioned identified points are compared to find the minimum value. The waveform data is corrected based on the minimum value discovered. Based on the corrected waveform data, the film thickness on the surface of the substrate is calculated. A substrate polishing apparatus configured as follows.
2. The substrate polishing apparatus according to claim 1, wherein correcting the waveform data includes applying the discovered minimum value at each point on the waveform data.
3. The substrate polishing apparatus according to claim 1 or 2, wherein the waveform data consists of a series of output signals obtained from the eddy current sensor when the eddy current sensor passes through a trajectory on the substrate as the polishing table and the polishing head rotate.
4. The substrate polishing apparatus according to claim 1 or 2, wherein the substrate is a substrate having one or more metal structures locally present on its surface or inside.
5. The substrate polishing apparatus according to claim 4, wherein the metal structure is a through electrode or metal wiring formed on the substrate.
6. The system further comprises an airbag capable of adjusting the polishing pressure on the substrate, The control unit is further configured to control the internal pressure of the airbag based on the calculated film thickness of the substrate. A substrate polishing apparatus according to claim 1 or 2.
7. A polishing table equipped with an eddy current sensor, comprising a polishing table configured to be rotatable, A polishing head that is rotatably configured to face the polishing table, and to which a substrate can be attached to the surface facing the polishing table, Control unit and A substrate polishing apparatus comprising, The control unit, A plurality of waveform data representing the output signal of the eddy current sensor when the eddy current sensor passes through a plurality of trajectories on the substrate, wherein a plurality of waveform data is acquired, each waveform data corresponding to one trajectory. Based on the comparison of the similarity of the plurality of waveform data, it is determined whether or not there is a metal structure on the track of the substrate. The waveform data corresponding to the trajectory in which the metal structure is determined to be present is the metal structure The signal value at the location where an object is detected is corrected to decrease, Based on the corrected waveform data and the waveform data other than the corrected waveform data, the film thickness on the surface of the substrate is calculated. A substrate polishing apparatus configured as follows.
8. The substrate polishing apparatus according to claim 7, wherein the determination includes determining that if a predetermined number of waveform data among the plurality of waveform data have a high degree of similarity and the remaining waveform data have a low degree of similarity, then it is determined that there is a metal structure on the trajectory corresponding to the remaining waveform data.
9. The substrate polishing apparatus according to claim 8, wherein the predetermined number is a majority.
10. A polishing table equipped with an eddy current sensor, comprising a polishing table configured to be rotatable, A polishing head that is rotatably configured to face the polishing table, and to which a substrate can be attached to the surface facing the polishing table, A method for calculating film thickness in a substrate polishing apparatus, comprising: The steps include acquiring waveform data of the output signal of the eddy current sensor when the eddy current sensor passes through a plurality of tracks on the substrate, The steps include identifying a plurality of corresponding points in the plurality of tracks on the substrate that are equidistant from the center of the substrate, The steps include comparing the values of the output signals of the eddy current sensors at the specified multiple points to find the minimum value, The steps include correcting the waveform data based on the minimum value discovered, A step of calculating the film thickness on the surface of the substrate based on the corrected waveform data, A method that includes this.