Control method for substrate processing

Abstract

To accurately detect an end point of substrate polishing using an acoustic sensor.SOLUTION: A control method for substrate processing in a substrate processing apparatus for performing polishing of a substrate by pressing the substrate against the polishing pad, includes: a step S10 of detecting an acoustic event occurring with polishing of the substrate and outputting the acoustic event as acoustic signals; a step 12 of generating power spectra from the acoustic signals, each of the power spectra indicating a spectrum of sound-pressure level; a step 13 of generating a power spectrum map indicating temporal change in the power spectra by arranging the power spectra in a time-series order; and a step S14 of detecting a polishing end point of the substrate based on the change in the sound-pressure level in the power spectrum map.SELECTED DRAWING: Figure 10

Description

Technical Field

[0001] The present invention relates to a control method for processing the surface of a substrate such as a semiconductor substrate.

Background Art

[0002] In the manufacturing process of semiconductor devices, polishing apparatuses for polishing the surface of a substrate such as a semiconductor substrate are widely used. In this type of polishing apparatus, the substrate rotates while being held by a substrate holding device called a top ring or a polishing head. In this state, while rotating the polishing table together with the polishing pad, the surface of the substrate is pressed against the polishing surface of the polishing pad, and the surface of the substrate is slidably contacted with the polishing surface in the presence of a polishing liquid, whereby the surface of the substrate is polished. When the film thickness of the substrate surface reaches a predetermined value by polishing the substrate surface, or when it is detected that an underlying layer (for example, a stopper layer) has appeared, the substrate polishing process is terminated.

[0003] In such a polishing process, it is required to accurately control the film thickness of the processed substrate surface, and for this purpose, it is important to accurately detect the end of substrate polishing. Various methods have been studied to detect the end of substrate polishing, and for example, it has also been proposed to detect a change in polishing sound using an acoustic sensor.

[0004] For example, in the control device described in Patent Document 1, the power spectrum of the polishing sound from the substrate is detected, and the S / N ratio per unit time is calculated from the change amount of the power spectrum, and when the obtained S / N ratio exceeds a threshold value, it is configured to determine that it is the end point of substrate polishing.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] In substrate polishing, polishing conditions (for example, the state of the polishing pad, the distribution of the polishing liquid, the pressing force by the polishing pad) are not always constant, and variations can occur in the amount of change in the power spectrum obtained by measurement with an acoustic sensor. Therefore, variations can occur in the timing when the value of the S / N ratio exceeds the threshold (the timing of the end of polishing). Also, when the S / N ratio does not exceed the threshold, the end of polishing cannot be detected.

[0007] The present invention has been made in view of the above, and an object thereof is to provide a substrate processing control method capable of accurately detecting the end point of substrate polishing using an acoustic sensor.

Means for Solving the Problems

[0008] One aspect of the present invention is a method for controlling substrate processing in a substrate processing apparatus that polishes a substrate by pressing the substrate against a polishing pad, the method including: detecting an acoustic event associated with the polishing of the substrate and outputting it as an acoustic signal; generating a power spectrum indicating a spectrum of the sound pressure level from the acoustic signal; generating a power spectrum map indicating the temporal change of the power spectrum by arranging the power spectra in time series; and detecting the end point of the polishing of the substrate based on the change in the sound pressure level in the power spectrum map.

[0009] In the step of detecting the end point of the polishing of the substrate, it is preferable to detect the change in the sound pressure level in the power spectrum map only in a predetermined monitoring frequency band. Thereby, the processing required for detecting the end of polishing of the substrate can be reduced. Also, in the step of detecting the end point of the polishing of the substrate, it is preferable that the end point determination unit sets the monitoring frequency band according to the material constituting each layer of the substrate. Thereby, an appropriate monitoring frequency band can be set according to the material constituting the substrate.

[0010] In the step of generating the power spectrum, it is preferable to generate the power spectrum using only the acoustic signal of a predetermined time period immediately before. Thereby, the process of generating the power spectrum can be reduced.

[0011] In the step of detecting the polishing end point of the substrate, when the polishing end index obtained by inputting the image of the power spectrum map into the learned model for generating the polishing end index indicating the degree of polishing completion exceeds a predetermined value, the polishing end point of the substrate can be detected. Thereby, it becomes possible to accurately detect the end point of substrate polishing.

[0012] In the above-described method for controlling substrate processing, a polishing head that forms a plurality of pressure chambers for pressing the substrate and a pressure control unit for performing pressure feedback control by individually controlling the pressures in the plurality of pressure chambers are provided in the substrate processing apparatus. In the step of detecting the polishing end point of the substrate, the time when a change occurs in the power spectrum map generated by each of the plurality of acoustic sensors provided in the polishing pad is detected, and the location where the surface of the substrate is exposed is specified from the difference in the time when the change occurs. The pressure control unit preferably reduces the pressure in the pressure chamber corresponding to the location where the surface of the substrate is exposed. Thereby, variations in the polishing amount on the polished surface of the substrate can be suppressed.

Advantages of the Invention

[0013] According to the present invention, since the power spectrum indicating the spectrum of the sound pressure level of the substrate polishing sound is generated and the polishing end point of the substrate is detected based on the change in the sound pressure level in the power spectrum map indicating the time change of the power spectrum, the end point of substrate polishing can be accurately detected using the acoustic sensor.

Brief Description of the Drawings

[0014]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Embodiments for Carrying Out the Invention

[0015] Hereinafter, a control method for substrate processing according to an embodiment of the present invention will be described with reference to the drawings. Note that the same or corresponding components are denoted by the same reference numerals and redundant descriptions are omitted.

[0016] FIG. 1 is a plan view showing the overall configuration of a substrate processing apparatus. The substrate processing apparatus 10 is partitioned into a load / unload unit 12, a polishing unit 13, and a cleaning unit 14, which are provided inside a rectangular housing 11. The substrate processing apparatus 10 also includes a control device 15 that controls the operations of processes such as substrate transfer, polishing, and cleaning.

[0017] The load / unload unit 12 includes a plurality of front load units 20, a traveling mechanism 21, and two transfer robots 22. Substrate cassettes for storing a large number of substrates (wafers) W are placed on the front load units 20. The transfer robot 22 has two hands vertically and moves on the traveling mechanism 21 to take out the substrate W from the substrate cassette in the front load unit 20 and send it to the polishing unit 13, and to return the processed substrate sent from the cleaning unit 14 to the substrate cassette.

[0018] The polishing unit 13 is an area for polishing (planarization processing) the substrate, and is provided with a plurality of polishing units 13A to 13D, which are arranged along the longitudinal direction of the substrate processing apparatus. Each polishing unit includes a top ring for polishing the substrate W on the polishing table while pressing it against the polishing pad, a liquid supply nozzle for supplying a liquid such as polishing liquid or pure water to the polishing pad, a dresser for dressing the polishing surface of the polishing pad, and an atomizer for spraying a mixed fluid of liquid and gas or a mist-like liquid onto the polishing surface to wash away polishing debris and abrasive grains remaining on the polishing surface.

[0019] Between the polishing unit 13 and the cleaning unit 14, first and second linear transporters 16 and 17 are provided as a transfer mechanism for transferring the substrate W. The first linear transporter 16 is movable between a first position for receiving the substrate W from the load / unload unit 12, second and third positions for transferring the substrate W between the polishing units 13A and 13B, and a fourth position for transferring the substrate W to the second linear transporter 17.

[0020] The second linear transporter 17 is movable between a fifth position for receiving the substrate W from the first linear transporter 16 and sixth and seventh positions for transferring the substrate W to and from the polishing units 13C and 13D. Between these transporters 16 and 17, a swing transporter 23 is provided for sending the substrate W from the fourth or fifth position to the cleaning unit 14 and from the fourth position to the fifth position.

[0021] The cleaning unit 14 includes a first substrate cleaning device 30, a second substrate cleaning device 31, a substrate drying device 32, and transfer robots 33 and 34 for transferring the substrate between these devices. The substrate W polished by the polishing unit is cleaned (primary cleaning) by the first substrate cleaning device 30 and then further cleaned (finish cleaning) by the second substrate cleaning device 31. The cleaned substrate is carried into the substrate drying device 32 from the second substrate cleaning device 31 and spin-dried. The dried substrate W is returned to the load / unload unit 12.

[0022] FIG. 2 is a perspective view schematically showing the configuration of the polishing unit. The polishing unit 40 includes a top ring (substrate holding device) 41 for holding and rotating the substrate (wafer) W, a polishing table 43 for supporting the polishing pad 42, and a polishing liquid supply nozzle 45 for supplying slurry (polishing liquid) to the polishing pad 42. Further, acoustic sensors 50 and 51 shown in FIG. 3 are provided below the polishing pad 42.

[0023] The top ring 41 is rotatably supported by a top ring shaft 47 and a top ring head cover 46, and is further configured to hold the substrate W by vacuum suction on its lower surface. Further, the top ring head cover 46 is pivotally directed by a rotation shaft 46a, and by the rotation of the rotation shaft 46a, the top ring 41 moves between a polishing position for polishing the substrate W and an exchange position for exchanging the substrate W.

[0024] The polishing table 43 is rotatable about a table axis 43a by a motor (not shown). The top ring 41 and the polishing table 43 rotate in the direction indicated by the arrow. In this state, the top ring 41 presses the substrate W against the polishing surface 42a on the upper side of the polishing pad 42 held by the polishing table 43. The substrate W is slidably contacted with the polishing pad 42 and polished in the presence of the polishing liquid supplied onto the polishing pad 42 from the polishing liquid supply nozzle 45.

[0025] The substrate W is composed of an upper layer such as a metal or a silicon oxide film and a lower layer such as a silicon film. Since the materials constituting the upper layer and the lower layer of the substrate W are different, when the polishing of the upper layer of the substrate W progresses and the lower layer is exposed, the acoustic spectrum (power spectrum) from the substrate W pressed against the polishing pad 42 changes. Note that the configuration of the substrate W in the present invention is not limited to this, and various materials used in the semiconductor chip manufacturing process can be used.

[0026] FIG. 3 is a side view schematically showing the configuration of the polishing unit. The top ring shaft 47 is connected to a polishing head motor 49 via a connecting means 48 such as a belt and is configured to be rotatable. By the rotation of the top ring shaft 47, the top ring 41 rotates in the direction indicated by the arrow. These connecting means 48 and the polishing head motor 49 are disposed inside the top ring head cover 46 of FIG. 2.

[0027] The acoustic sensors 50 and 51 are general acoustic emission sensors (AE sensors), and are arranged in two in the radial direction of the polishing pad 42 and disposed below the polishing pad 42. When the substrate W being polished is pressed against the polishing pad 42 and undergoes deformation, strain energy is released from the substrate W as elastic waves (AE waves). The acoustic sensors 50 and 51 detect the elastic waves transmitted through the polishing pad 42 and output them as electrical signals (acoustic signals). Alternatively, the acoustic sensors 50 and 51 may be composed of ultrasonic microphones, and detect the polishing sound caused by the friction between the substrate W pressed by the top ring 41 and the polishing pad 42 and output it as an electrical signal (acoustic signal). The output acoustic signal is connected to a rotary connector 61 installed inside the table shaft 43a via a connector attached to the side surface of the table shaft 43a. The rotary connector 61 is connected to the control device 15, and the acoustic signal corresponding to the polishing sound of the substrate W is sent to the control device 15. Thereby, the acoustic signal from the acoustic sensors 50 and 51 can be output to the control device without being affected by the rotation of the table shaft 43a.

[0028] FIG. 4 is an explanatory view when the polishing table 43 is viewed from the bottom. Depressions 43b and 43c are formed on the bottom surface of the polishing table 43, and each of the acoustic sensors 50 and 51 is fixed to the polishing table 43 in a state of entering the depressions 43b and 43c. By fixing the acoustic sensors 50 and 51 inside the polishing table 43 (close to the polishing surface), the detection accuracy by the acoustic sensors 50 and 51 can be improved.

[0029] FIG. 5 shows an example of the configuration of the control device 15. The control device 15 is, for example, a general-purpose computer device, and includes a CPU, a memory for storing a control program, an input unit, a display unit, and the like. By starting the control program stored in the memory, the control device 15 operates as a polishing control unit 52, a spectrum generation unit 54, a color map update unit 56, and an end point determination unit 58, thereby comprehensively controlling the operation of the polishing unit 40. Note that the configuration of the control device 15 is not limited to that shown in FIG. 5, and it also has a configuration for controlling the operations of other elements of the substrate processing apparatus 10 (for example, the load / unload unit 12 and the cleaning unit 14).

[0030] The control program for controlling the operation of the substrate processing apparatus 10 may be installed in advance in the computer constituting the control device 15, or may be stored in a storage medium such as a CD-ROM or a DVD-ROM. Further, it may be installed in the control device 15 via the Internet.

[0031] The polishing control unit 52 controls the operations of the top ring 41, the polishing table 43, etc. constituting the polishing unit 40, and performs a polishing process on the substrate W held by the top ring 41.

[0032] The spectrum generation unit 54 performs FFT (Fast Fourier Transform) on the data of the acoustic signals (signals resulting from the distortion of the substrate W pressed by the polishing pad 42) transmitted from the acoustic sensors 50 and 51, extracts the frequency components and their intensities, and outputs them as the power spectrum (sound pressure level with respect to frequency) of the acoustic signals of the substrate W. Here, regarding the number of data of the acoustic signals used for generating the power spectrum, all the data acquired since the start of substrate polishing may be used, but it is desirable to use only the data of the acoustic signals for a certain recent period (for example, 10 seconds), thereby shortening the time for the power spectrum generation process.

[0033] FIG. 6 is a graph showing an example of signals transmitted from the acoustic sensors 50 and 51. The horizontal axis represents the elapsed time since the start of substrate polishing, and the vertical axis represents the intensity (voltage) of the acoustic signal. As the substrate W is polished, a signal (acoustic signal) is generated due to the distortion of the substrate W pressed by the top ring 41. The spectrum generation unit 54 generates a power spectrum using the signal within the most recent, for example, 10 seconds (the signal within the section included in the “analysis window” indicated by the dotted line in FIG. 6). In the present embodiment, a power spectrum may be generated using only the signal from one of these two acoustic sensors 50 and 51, or an average value may be used. Further, the power spectrum based on the acoustic signal from the acoustic sensor 50 and the power spectrum based on the acoustic signal from the other acoustic sensor 51 may be generated separately and each may be configured to be used for the determination of end point detection described later.

[0034] FIG. 7 is a graph showing an example of the power spectrum generated as described above (here, showing the case where only the acoustic signal of one of the two acoustic sensors 50 and 51 is used). The horizontal axis represents the frequency, and the vertical axis represents the sound pressure level. As described above, the spectrum generation unit 54 generates a power spectrum using the acoustic signal included in the analysis window (see FIG. 6) at regular time intervals (for example, at 1-second intervals). As a result, as the substrate W is polished, data of a plurality of power spectra are generated in time series (in FIG. 7, the generation of graphs for each analysis window is schematically shown by stacking three).

[0035] Note that since the sound pressure level in the low-frequency region is often irrelevant to the change in the substrate polishing situation, a high-pass filter (or band-pass filter) may be provided on the output side of the acoustic sensors 50 and 51 to cut the signal in the low-frequency region.

[0036] The color map update unit 56 generates a graph (color map) showing the temporal changes in frequency and sound pressure level by arranging the data of the power spectrum generated by the spectrum generation unit 54 in time series. FIG. 8 is a graph showing an example of the color map, where the horizontal axis represents time and the vertical axis represents frequency, and the sound pressure level at a certain point in time and frequency is configured by color coding (or by the distribution of black and white densities). The generated color map is displayed on a display unit (display device) provided in the control device 15.

[0037] In the example of FIG. 8, the color map is configured such that the sound pressure level is color-coded and displayed at predetermined values (for example, 20 dB), but it is not limited to this mode. For example, the color map may be configured such that the color changes in a gradient manner.

[0038] In the graph of FIG. 8, "0" on the horizontal axis (time) represents the start point of polishing (that is, the point in time when the measurement of the sound pressure signal by the acoustic sensors 50 and 51 is started). Since the spectrum generation unit 54 generates the power spectrum using the signal in the most recent, for example, 10 seconds (the time corresponding to the width of the "analysis window" in FIG. 5), the power spectrum in the first approximately 10 seconds (when the signal has not been generated) is not used for the polishing end determination described later. Alternatively, it may be configured not to generate the power spectrum. In the example of FIG. 8, it is shown that the sound pressure level in the region with a low frequency is relatively high, and the sound pressure level decreases as the frequency increases.

[0039] The end point determination unit 58 monitors the sound pressure level of the color map in a predetermined frequency band (monitoring range) and determines whether a change has occurred in the color map in the monitoring range. In the example of FIG. 7, at the point in time when 40 seconds have elapsed since the start of polishing, the sound pressure level in the 12 - 16 kHz band has increased. This is due to the fact that the lower layer, which was hidden by the upper layer at the start of polishing, is gradually exposed, and the spectrum of the acoustic signal from the substrate W has changed under the influence of the underlying layer.

[0040] When the end point determination unit 58 detects a change in the color map within the monitoring range, it sends a signal instructing the end of substrate polishing to the polishing control unit 52. For example, when the rate of change in the sound pressure level over a certain period exceeds a predetermined value, when the area of the region where the sound pressure level increases in the color map exceeds a predetermined value, when the sound pressure level within the monitoring range increases and then decreases and the amount of fluctuation becomes less than the threshold value, it can be detected that the lower layer of the substrate W is exposed.

[0041] The monitoring range for monitoring the sound pressure level in the end point determination unit 58 can be set according to the combination of materials of each layer constituting the substrate W. Alternatively, prior to the actual polishing of the substrate W, test polishing may be performed using a dummy substrate having a common layer structure, and the frequency band in which a change occurs in the generated color map may be set as the monitoring range.

[0042] The storage unit 60 is, for example, a non-volatile memory device, and stores information such as information on signals received from the acoustic sensors 50 and 51, information on the power spectrum generated by the spectrum generation unit 54, information on the color map generated by the color map update unit 56, and the monitoring range determined for each type of layer constituting the substrate W, and is read out as appropriate.

[0043] As shown in FIG. 9, the top ring 41 includes a head body 70 fixed to the lower end of the top ring shaft 47, a retainer ring 71 that supports the side edge of the substrate W, and a flexible elastic film 72 that presses the substrate W against the polishing surface of the polishing pad 42. The retainer ring 71 is arranged so as to surround the substrate W and is connected to the head body 70. The elastic film 72 is attached to the head body 70 so as to cover the lower surface of the head body 70.

[0044] The head body 70 is formed of a resin such as engineering plastic (e.g., PEEK), and the elastic film 72 is formed of a rubber material having excellent strength and durability such as ethylene propylene rubber (EPDM), polyurethane rubber, or silicone rubber.

[0045] The top ring body 70 and the retainer ring 71 that constitute the top ring 41 are configured to rotate integrally with the rotation of the top ring shaft 47.

[0046] The retainer ring 71 is disposed so as to surround the top ring body 70 and the elastic film 72. This retainer ring 71 is a member made of a ring-shaped resin material that contacts the polishing surface 42a of the polishing pad 42, and is disposed so as to surround the outer peripheral edge of the substrate W held by the top ring body 70, and supports the outer peripheral edge of the substrate W so that the substrate W during polishing does not jump out of the top ring 41.

[0047] The upper surface of the retainer ring 71 is connected to an annular retainer ring pressing mechanism (not shown), and a uniform downward load is applied to the entire upper surface of the retainer ring 71. Thereby, the lower surface of the retainer ring 71 is pressed against the polishing surface 42a of the polishing pad 42.

[0048] The elastic film 72 is provided with a plurality (four in FIG. 9) of concentrically arranged annular peripheral walls 72a, 72b, 72c, 72d. By these plurality of peripheral walls 72a to 72d, a circular first pressure chamber D1 located at the center and annular second, third, and fourth pressure chambers D2, D3, D4 are formed between the upper surface of the elastic film 72 and the lower surface of the head body 70.

[0049] In the head body 70, a flow path G1 communicating with the central first pressure chamber D1 and flow paths G2 to G4 communicating with the second to fourth pressure chambers are respectively formed. These flow paths G1 to G4 are respectively connected to a fluid supply source 74 via fluid lines. Open / close valves V1 to V4 and a pressure controller (not shown) are installed in the fluid lines.

[0050] Immediately above the retainer ring 71, a retainer chamber D5 is formed. The retainer pressure chamber D5 is connected to a fluid supply source 74 via a flow path G5 formed in the head body 70, an on-off valve V5, and a fluid line in which a pressure controller (not shown) is installed. The pressure controllers installed in the fluid line each have a pressure adjustment function for adjusting the pressure of the pressure fluid supplied from the fluid supply source 74 to the pressure chambers D1 to D4 and the retainer pressure chamber D5. The pressure controllers and the on-off valves V1 to V5 are controlled by the control device 15 for their operations.

[0051] Hereinafter, the operation of the substrate polishing apparatus 10 having the above configuration will be described with reference to the flowchart of FIG. 10. When polishing of the substrate W is started, the acoustic sensors 50 and 51 detect the polishing sound of the substrate W transmitted through the polishing pad 42, convert it into an acoustic signal indicating the sound pressure level, and output it to the control device 15 (step S10).

[0052] The control device 15 stores the data of the acoustic signals received from the acoustic sensors 50 and 51 in the storage unit 60. Then, the control device 15 determines whether the data amount of the acoustic signals stored in the storage unit 60 exceeds a predetermined value (for example, the data amount corresponding to 10 seconds) (step S11). If the data amount exceeds the predetermined value, the spectrum generation unit 54 reads out the data of the acoustic signals for the most recent 10 seconds stored in the storage unit 60, and generates a frequency spectrum (power spectrum) at a certain point in time by performing FFT processing (step S12). The data of the frequency spectrum is stored in the storage unit 60.

[0053] Next, the color map update unit 56 of the control device 15 arranges the data of the frequency spectrum stored in the storage unit 60 in time series, generates, for example, a color map as shown in FIG. 8, and updates it (step S13). The data of the color map is stored in the storage unit 60.

[0054] The end point determination unit 58 determines whether a predetermined end point detection condition is satisfied for the color map generated (updated) by the color map update unit 56 (for example, whether a predetermined fluctuation has occurred in the sound pressure level in the monitor area (monitor frequency area)) (step S14). If the end point detection condition is not satisfied, the control device 15 receives the data of the acoustic signal from the acoustic sensors 50 and 51 (step S15), returns to step S12, generates a power spectrum in the spectrum generation unit 54, and updates the color map in the color map update unit 56.

[0055] On the other hand, when it is determined in the end point determination unit 58 that the end point detection condition is satisfied, the polishing control unit 52 stops the rotation of the top ring 41 and the polishing pad 42 and ends the polishing process (step S16).

[0056] In this way, based on the acoustic signal obtained by the acoustic sensor, a color map (intensity distribution diagram) of the sound pressure level is generated, and the end point of the substrate polishing is detected from the change in the color map, so that the end point of the substrate polishing can be accurately detected.

[0057] In the above embodiment, the power spectrum is generated using the acoustic signals from the two acoustic sensors 50 and 51. However, the acoustic sensors in the present invention are not limited to two, and one or three or more acoustic sensors may be used.

[0058] Also, the power spectrum and the color map may be individually generated using the acoustic signals acquired from each of the two acoustic sensors 50 and 51, and the substrate polishing may be ended when either or both of the color maps satisfy the end point detection condition. In this case, the location where the surface of the substrate W is exposed (the sound source in FIG. 11) is specified from the time difference at which changes occur in the two color maps, and the pressure in the pressure chamber of the area corresponding to the exposed location is decreased to adjust the polishing rate of the exposed location. Thereby, the variation in the film thickness distribution in the plane of the substrate during polishing can be suppressed.

[0059] In the above embodiment, an acoustic sensor embedded in the polishing table is used to generate an acoustic signal of the substrate W. However, the present invention is not limited to this. For example, as shown in FIG. 12, a sound collecting microphone (ultrasonic microphone) 80 as an acoustic sensor for polishing sound is arranged above the polishing table, and an acoustic signal from the substrate W is generated using the sound collecting microphone 80, and a color map may be generated by the same procedure as in the above embodiment. In the example shown in FIG. 12, the sound collecting microphone 80 is fixed to the bottom of the top ring head cover 46 by a holding mechanism 82.

[0060] FIG. 13 is an example of a color map generated by an acoustic signal obtained by the sound collecting microphone 80. Similar to the case of using an acoustic sensor embedded in the polishing table, by detecting a change in sound pressure level in a predetermined frequency range (monitoring range), exposure of the lower layer (end of substrate polishing) can be detected. In the example of FIG. 13, the color map is configured such that the sound pressure level is color-coded and displayed for each predetermined value. However, the present invention is not limited to this aspect. For example, the color map may be configured such that the color changes in a gradient manner.

[0061] In the above embodiment, the end point of substrate polishing is detected from the change in the color map. However, the method for detecting the end of substrate polishing is not limited to this. For example, a learned model may be generated by machine learning a plurality of color map images indicating that the end point has been reached, and end point detection may be performed by image detection using the learned model.

[0062] FIG. 14 shows the configuration of a system in an embodiment for performing image detection using a learned model. For the same components as those in the above-described embodiment, the same reference numerals are given and detailed description is omitted. In FIG. 14, the system includes a control device 100 that performs substrate polishing and end point detection, and a learning device 110 that performs machine learning on images of the color map. The control device 100 includes an end determination unit 102 and an image extraction unit 104 in addition to the configuration of the control device 15 described above.

[0063] The end determination unit 102 includes a learned model 106 described later. This learned model 106 is a learned machine learning model that has learned, for example, using a neural network, to estimate the degree to which the generated color map image matches the polished end image. This learned model 106 is stored in the storage unit 60 of the control device 100 from the learning device 110, and is read out by the end determination unit 54 when the control device 100 performs polished end determination by image detection.

[0064] As this neural network, for example, the convolutional neural network 120 shown in FIG. 15 is used. The convolutional neural network 120 has a structure in which convolutional layers 122 and pooling layers 124 are alternately connected, the output of the pooling layer 124 on the output side is input to the fully connected layer 126, and the output of the fully connected layer 126 is input to the output layer 128.

[0065] In the convolutional layer 122, by calculating the correlation between the image data of the input image and a predetermined weight filter, feature amounts in each local region of the input image are output. In the pooling layer 124, for the feature amounts in the local regions output by the convolutional layer 122, the maximum value or the average value is output. The fully connected layer 126 is composed of a plurality of layers, each layer includes one or a plurality of neurons (nodes), and the neurons of adjacent layers are connected to each other. The output layer 128 is arranged on the most output side of the neural network 120, and outputs estimation information indicating the degree to which the input color map image matches the polished end image.

[0066] Weights are set for the connection of neurons, and thresholds are set for each neuron. The output of each neuron is determined by whether the sum of the product of the input to each neuron and the weight exceeds the threshold, and thereby estimation information is output in the neural network. When the value of the estimation information output from the learned model exceeds a preset reference value, the polishing determination unit determines that the input image matches the polished end image and ends the substrate polishing.

[0067] Note that the neural network in this embodiment is not limited to this. For example, a fully-connected neural network having an input layer, an intermediate layer, and an output layer may be used, or a convolutional neural network and a fully-connected neural network may be combined and used. Further, a recurrent neural network having a loop inside (for example, an LSTM network) may be provided.

[0068] The image extraction unit 104 extracts a part of the image of the color map determined at a predetermined frequency band and a predetermined time from the color map updated by the color map update unit 56, and inputs this to the learned model 102 of the end determination unit. Thereby, the image data of the portion unnecessary for end point detection is omitted, and the processing time for end point detection by image detection can be shortened. Further, in the image extraction unit 104, the resolution of the extracted image may be reduced, and this can also shorten the processing time for end point detection.

[0069] The learning device 110 is, for example, a general-purpose computer, and includes a CPU, a memory for storing a learning program, an input device, a display device, etc., and is connected to the control device 100 via a communication line (not shown). The learning device 110 operates as an image input unit 112, a teacher data storage unit 114, a learning unit 116, and a learned model storage unit 118 by starting a learning program stored in advance in a memory (not shown) (or installed through a network). Note that the learning device 110 and the control device 100 may be integrally configured.

[0070] The image input unit 112 inputs an image of a color map at the time when the substrate polishing is completed by test polishing, and stores it in the teacher data storage unit 114 as a part of the image (teacher data) determined within a predetermined frequency band and a predetermined time. The learning unit 116 has a configuration equivalent to that of the neural network 120 described above, and adjusts the weights and thresholds of each neuron so that estimation information exceeding a reference value can be obtained as an output when the teacher data is input, thereby performing learning of the neural network. When estimation information exceeding the reference value is output for a plurality of pieces of teacher data stored in the teacher data storage unit 114, the learning is terminated and stored in the learned model storage unit 118 as a learned model. Further, the learning device 110 sends the data of the learned model for which the learning has been completed to the control device 100, whereby the learned model 106 of the control device 100 is updated.

[0071] FIG. 16 is a flowchart schematically showing a method of manufacturing a semiconductor device including a method of controlling substrate processing according to the present embodiment. First, a substrate W is prepared (step S101). Next, an opening pattern is formed on the surface of the substrate W using, for example, photolithography (step S102), and a film such as a metal or a silicon oxide film is formed on the surface of the substrate W having the opening pattern using, for example, chemical vapor deposition (CVD) or physical vapor deposition (PVD) (step S103). Then, polishing of the surface of the substrate W is performed by the method of controlling substrate processing according to the present embodiment (step S104). The formation of the opening pattern and film formation on the surface of the substrate W and the polishing of the substrate W may each be performed a plurality of times.

[0072] The above-described embodiments are described for the purpose of enabling a person having ordinary knowledge in the technical field to which the present invention pertains to implement the present invention. Various modifications of the above embodiments can be naturally made by those skilled in the art, and the technical idea of the present invention can also be applied to other embodiments. The present invention is not limited to the described embodiments, but is construed in the broadest scope in accordance with the technical idea defined by the claims.

Explanation of reference numerals

[0073] 10 Substrate processing apparatus 15, 100 Control device 40 Grinding unit 41 Top ring 42 Grinding pad 50, 51 Acoustic sensor 52 Grinding control unit 54 Spectrum generation unit 56 Color map update unit 58, 102 End point determination unit 60 Memory unit 80 Sound collection microphone 106 Trained model 110 Learning device W Substrate

Claims

1. A method for controlling substrate processing in a substrate processing apparatus that polishes a substrate by pressing the substrate against a polishing pad, comprising: detecting an acoustic event associated with polishing of the substrate and outputting it as an acoustic signal; generating a power spectrum showing a spectrum of sound pressure levels from the acoustic signal; generating a power spectrum map showing a temporal change of the power spectrum by arranging the power spectra in time series; detecting an end point of polishing of the substrate based on a change in the sound pressure level in the power spectrum map; wherein the change in the sound pressure level is a change over time of the power spectrum for each frequency of the acoustic signal; the step of detecting the end point of polishing of the substrate is: detecting the end point of polishing of the substrate when a rate of change of the sound pressure level over a certain time exceeds a predetermined value; detecting the end point of polishing of the substrate when an area of a region where the sound pressure level has increased in the power spectrum map exceeds a predetermined value, or detecting the end point of polishing of the substrate when a fluctuation amount of the sound pressure level becomes less than a threshold value when the sound pressure level in a predetermined frequency band increases and then decreases; wherein the predetermined frequency band is determined based on a power spectrum map generated from a test polish performed using a dummy substrate having the same layer structure as the substrate. A control method characterized by that.

2. The control method according to claim 1, wherein in the step of detecting the end point of polishing of the substrate, the frequency band is set according to a material constituting each layer of the substrate.

3. The control method according to claim 1 or 2, wherein in the step of generating the power spectrum, the power spectrum is generated using only the acoustic signal of a predetermined time immediately before generating the power spectrum.

4. The substrate processing apparatus includes a polishing head that forms a plurality of pressure chambers for pressing the substrate, and a pressure control unit for performing pressure feedback control by individually controlling the pressures in the plurality of pressure chambers. In the step of detecting the polishing end point of the substrate, the time when a change occurs in the power spectrum map generated by each of a plurality of acoustic sensors provided in the polishing pad is detected, and a portion where a lower layer hidden under the upper layer of the substrate is exposed is specified from the difference in the time when the change occurs. The control method according to any one of claims 1 to 3, wherein the pressure control unit reduces the pressure in the pressure chamber corresponding to the portion where the lower layer of the substrate is exposed.

5. The control method according to claim 1, wherein an acoustic sensor that detects an acoustic event associated with polishing of the substrate is disposed in a depression formed in a polishing table that supports the polishing pad.

6. A method for controlling substrate processing in a substrate processing apparatus that polishes a substrate by pressing the substrate against a polishing pad, detecting an acoustic event associated with polishing of the substrate and outputting it as an acoustic signal; generating a power spectrum indicating a spectrum of sound pressure levels from the acoustic signal; generating a power spectrum map showing a temporal change of the power spectrum by arranging the power spectra in time series; detecting a polishing end point of the substrate based on a change in the sound pressure level in the power spectrum map, and comprising: The change in the sound pressure level is a change in the power spectrum over time for each frequency of the acoustic signal. In the step of detecting the polishing end point of the substrate, when the polishing end index obtained by inputting an image of the power spectrum map to a learned model for generating a polishing end index indicating the degree of polishing completion exceeds a predetermined value, the polishing end point of the substrate is detected. A control method characterized by this.

7. preparing a substrate; forming an opening pattern on the surface of the substrate; forming a film on the surface of the substrate; A method for manufacturing a semiconductor device, comprising pressing a polishing pad against the surface of the substrate by a substrate processing apparatus having the polishing pad to perform polishing, The step of performing the polishing is detecting an acoustic event associated with polishing of the substrate and outputting it as an acoustic signal; generating a power spectrum indicating a spectrum of sound pressure levels from the acoustic signal; Generating a power spectrum map showing the temporal change of the power spectrum by arranging the power spectra in time series; Detecting the polishing end point of the substrate based on the change in the sound pressure level in the power spectrum map; The change in the sound pressure level is a change in the power spectrum over time for each frequency of the acoustic signal; The step of detecting the polishing end point of the substrate; Detecting the polishing end point of the substrate when the rate of change of the sound pressure level over a certain period of time exceeds a predetermined value; Detecting the polishing end point of the substrate when the area of the region where the sound pressure level increases in the power spectrum map exceeds a predetermined value, or Detecting the polishing end point of the substrate when the amount of fluctuation of the sound pressure level becomes less than a threshold value when the sound pressure level increases and then decreases in a predetermined frequency band; The manufacturing method is characterized in that the predetermined frequency band is determined based on a power spectrum map generated from test polishing performed using a dummy substrate.

Images