Substrate Processing Method and Substrate Processing Apparatus

Abstract

To provide a substrate processing method and a substrate processing device, capable of setting a thickness of an object constituting a substrate at a desired thickness.SOLUTION: A substrate processing method processes a substrate W of a processing object by supplying a process liquid from a first nozzle 41 to the substrate W of the processing object. The substrate processing method includes steps of: acquiring preprocessing surface information on a surface of the substrate W of the processing object (step S202); selecting one learned model 133 from a plurality of learned models 133 on the basis of the preprocessing surface information (step S203); and processing the substrate W of the processing object on the speed condition obtainable on the basis of the preprocessing surface information and the selected learned model 133 (step S205). The learned model 133 outputs the speed condition, which is a movement speed relative to the first nozzle 41 and the substrate W, by receiving input of the preprocessing surface in formation.SELECTED DRAWING: Figure 15

Description

Technical Field

[0001] The present invention relates to a substrate processing method and a substrate processing apparatus.

Background Art

[0002] There is known a substrate processing apparatus that processes a wafer having a film formed on its surface and adjusts the film thickness of the film and removes foreign matter by liquid processing. As one type of such a substrate processing apparatus, there is a single-wafer type substrate processing apparatus provided with a nozzle for supplying a processing liquid for etching to the wafer surface (see, for example, Patent Document 1). When executing the etching process, the substrate processing apparatus of Patent Document 1 discharges the processing liquid for etching from the nozzle while moving the nozzle based on a previously defined speed profile.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in a substrate processing apparatus that moves a nozzle based on a previously defined speed profile, when performing an etching process on a plurality of wafers having different surface shapes (profiles) under the same conditions, the variation in the surface shape of the processed wafers may increase. Specifically, the variation in the thickness of the insulating film or the like on the surface of each wafer increases. That is, it may be difficult to make the thickness of the object constituting the substrate the desired thickness.

[0005] The present invention has been made in view of the above problems, and an object thereof is to provide a substrate processing method and a substrate processing apparatus capable of making the thickness of the object constituting the substrate the desired thickness.

Means for Solving the Problems

[0006] According to one aspect of the present invention, a substrate processing method supplies a processing liquid from a nozzle to a substrate to be processed and Etching executes processing on the substrate to be processed. The substrate processing method includes a step of acquiring pre-processing surface information regarding the surface of the substrate to be processed, a step of selecting one learned model from a plurality of learned models based on the pre-processing surface information, and a step of processing the substrate to be processed under a speed condition obtained based on the pre-processing surface information and the selected One learned model. The Etching learned model outputs the speed condition, which is the relative movement speed between the nozzle and the substrate, when the pre-processing surface information is input. One

[0007] In one aspect of the present invention, the plurality of learned models cluster processing amount information regarding the processing amount when processing a plurality of substrates to be learned under a plurality of speed conditions into a plurality of clusters, and each cluster Speed of degree condition and Treatment location of is a plurality of models obtained by performing machine learning using the processing amount information.

[0008] In one aspect of the present invention, clusters in which the number of substrates to be learned is equal to or greater than a predetermined number are selected from the plurality of clusters, and the selected clusters Speed of degree condition and Treatment location of are used to perform machine learning using the processing amount information, thereby obtaining the plurality of learned models.

[0009] In one aspect of the present invention, processing amount information having a correlation with other processing amount information in each cluster that is equal to or higher than a predetermined condition is selected, and the selected processing amount information is used to perform machine learning, thereby obtaining the One learned model.

[0010] In one aspect of the present invention, in the step of selecting the one learned model, based on the pre - processing surface information, a target value of the processing amount for processing the substrate to be processed is derived, and from the plurality of learned models, one cluster having a correlation with the target value of the processing amount equal to or higher than a predetermined condition Treatment location of is selected, and the one learned model obtained using the processing amount information is selected.

[0011] In one aspect of the present invention, the pre - processing surface information includes information indicating the thickness distribution of the substrate to be processed.

[0012] A substrate processing apparatus according to one aspect of the present invention supplies a processing liquid to a substrate to be processed to Etching execute processing. The substrate processing apparatus includes a nozzle, an acquisition unit, and a control unit. The nozzle supplies the processing liquid to the substrate. The acquisition unit acquires pre - processing surface information regarding the surface of the substrate to be processed. The control unit selects one learned model from a plurality of learned models based on the pre - processing surface information. The control unit processes the substrate to be processed under a speed condition obtained based on the pre - processing surface information and the selected One learned model. The Etching learned model outputs the speed condition, which is the relative movement speed between the nozzle and the substrate, when the pre - processing surface information is input. One

Advantages of the Invention

[0013] According to the present invention, it is possible to provide a substrate processing method and a substrate processing apparatus capable of making the thickness of an object constituting the substrate a desired thickness.

Brief Description of the Drawings

[0014]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8A

Figure 8B

Figure 8C

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Mode for Carrying Out the Invention

[0015] Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the present invention is not limited to the following embodiments, and can be implemented in various aspects without departing from the gist thereof. Note that, in some cases, the description may be omitted for overlapping parts. Also, in the drawings, the same or corresponding parts are denoted by the same reference numerals and the description will not be repeated.

[0016] Referring to FIG. 1, a substrate processing apparatus 100 according to an embodiment of the present invention will be described. FIG. 1 is a schematic diagram of the substrate processing apparatus 100 of the present embodiment. Specifically, FIG. 1 is a schematic plan view of the substrate processing apparatus 100. The substrate processing apparatus 100 is a single-wafer type apparatus that processes substrates W one by one. In the present embodiment, the substrate W is a semiconductor wafer. The substrate W is substantially disc-shaped.

[0017] As shown in FIG. 1, the substrate processing apparatus 100 includes a plurality of processing units 1, a fluid cabinet 100A, a plurality of fluid boxes 100B, a plurality of load ports LP, an indexer robot IR, a center robot CR, and a control device 101.

[0018] Each of the load ports LP stores a plurality of substrates W stacked thereon. The indexer robot IR transports the substrate W between the load port LP and the center robot CR. The center robot CR transports the substrate W between the indexer robot IR and the processing unit 1. Each of the processing units 1 supplies a processing liquid to the substrate W and executes a process on the substrate W. The fluid cabinet 100A stores the processing liquid.

[0019] The plurality of processing units 1 form a plurality of towers TW (four towers TW in FIG. 1) arranged so as to surround the center robot CR in a plan view. Each tower TW includes a plurality of processing units 1 (three processing units 1 in FIG. 1) stacked vertically. The fluid boxes 100B respectively correspond to the plurality of towers TW. The processing liquid in the fluid cabinet 100A is supplied to all the processing units 1 included in the tower TW corresponding to the fluid box 100B via one of the fluid boxes 100B.

[0020] The control device 101 controls the operations of each part of the substrate processing device 100. For example, the control device 101 controls the load port LP, the indexer robot IR, and the center robot CR.

[0021] In the present embodiment, the control device 101 functions as a learning device. Specifically, the control device 101 performs machine learning. The machine learning is, for example, any one of supervised learning, unsupervised learning, semi-supervised learning, reinforcement learning, and deep learning.

[0022] Subsequently, with reference to FIG. 2, the processing unit 1 of the present embodiment will be described. FIG. 2 is a schematic diagram of the processing unit 1 of the present embodiment. Specifically, FIG. 2 is a schematic cross-sectional view of the processing unit 1.

[0023] As shown in FIG. 2, the processing unit 1 processes the object constituting the substrate W with a processing liquid. Hereinafter, the object to be processed by the processing liquid is referred to as "object TG". The object TG is, for example, a substrate body (for example, a substrate body made of silicon), or a substance formed on the surface of the substrate body. The substance formed on the surface of the substrate body is, for example, a substance of the same material as the substrate body (for example, a layer made of silicon), or a substance of a different material from the substrate body (for example, a silicon oxide film, a silicon nitride film, or a resist). The "substance" may constitute a film.

[0024] In the present embodiment, the processing liquid includes an etching liquid, and the processing unit 1 performs an etching process. The object TG is processed (etched) with the etching liquid. The etching liquid is a chemical solution. The etching liquid is, for example, hydrofluoric acid and nitric acid (a mixed solution of hydrofluoric acid (HF) and nitric acid (HNO3)), hydrofluoric acid, buffered hydrofluoric acid (BHF), ammonium fluoride, HFEG (a mixed solution of hydrofluoric acid and ethylene glycol), or phosphoric acid (H3PO4).

[0025] The processing unit 1 includes a chamber 2, a spin chuck 3, a spin motor unit 5, a nozzle moving mechanism 6, a measurement unit 8, a probe moving mechanism 9, a plurality of guards 10 (two guards 10 in FIG. 2), a first nozzle 41, and a second nozzle 71. Further, the substrate processing apparatus 100 includes an etching solution supply unit 4 and a rinse solution supply unit 7. The etching solution supply unit 4 has a first supply pipe 42, and the rinse solution supply unit 7 has a second supply pipe 72. Note that the measurement unit 8 is an example of the "acquisition unit" of the present invention. Also, the first nozzle 41 is an example of the "nozzle" of the present invention.

[0026] The chamber 2 has a substantially box shape. The chamber 2 houses a substrate W, a spin chuck 3, a spin motor unit 5, a nozzle moving mechanism 6, a plurality of guards 10, a measurement unit 8, a probe moving mechanism 9, a first nozzle 41, a second nozzle 71, a part of the first supply pipe 42, and a part of the second supply pipe 72.

[0027] The spin chuck 3 holds the substrate W horizontally. Specifically, the spin chuck 3 has a plurality of chuck members 32 and a spin base 33. The plurality of chuck members 32 are provided on the spin base 33 along the periphery of the substrate W. The plurality of chuck members 32 hold the substrate W in a horizontal posture. The spin base 33 is substantially disk-shaped and supports the plurality of chuck members 32 in a horizontal posture.

[0028] The spin motor unit 5 rotates the substrate W and the spin chuck 3 integrally about a first rotation axis AX1. The first rotation axis AX1 extends in the vertical direction. In the present embodiment, the first rotation axis AX1 extends in a substantially vertical direction. Specifically, the spin motor unit 5 rotates the spin base 33 about the first rotation axis AX1. Therefore, the spin base 33 rotates about the first rotation axis AX1. As a result, the substrate W held by the plurality of chuck members 32 provided on the spin base 33 rotates about the first rotation axis AX1.

[0029] Specifically, the spin motor unit 5 includes a motor body 51, a shaft 53, and an encoder 55. The shaft 53 is coupled to the spin base 33. The motor body 51 rotates the shaft 53. As a result, the spin base 33 rotates.

[0030] The encoder 55 measures the rotation speed of the substrate W. The encoder 55 generates a signal indicating the rotation speed of the substrate W. Specifically, the encoder 55 generates a rotation speed signal indicating the rotation speed of the motor body 51.

[0031] The first nozzle 41 supplies the etching solution to the substrate W. Specifically, the first nozzle 41 discharges the etching solution toward the rotating substrate W. The etching solution supply unit 4 supplies the etching solution to the first nozzle 41. Specifically, the first nozzle 41 is connected to one end of the first supply pipe 42. The etching solution is supplied to the first nozzle 41 through the first supply pipe 42. The first supply pipe 42 is a tubular member through which the etching solution flows.

[0032] The nozzle moving mechanism 6 moves the first nozzle 41. In the present embodiment, the nozzle moving mechanism 6 moves the first nozzle 41 in a substantially horizontal direction. Specifically, the nozzle moving mechanism 6 pivots the first nozzle 41 about a second rotation axis AX2 along a substantially vertical direction. The first nozzle 41 discharges the etching solution toward the substrate W while moving (while pivoting). The first nozzle 41 may be referred to as a scan nozzle.

[0033] Specifically, the nozzle movement mechanism 6 includes a nozzle arm 61, a first rotating shaft 63, and a first driving unit 65. The nozzle arm 61 extends along a substantially horizontal direction. The first nozzle 41 is disposed at the tip of the nozzle arm 61. The nozzle arm 61 is coupled to the first rotating shaft 63. The first rotating shaft 63 extends along a substantially vertical direction. The first driving unit 65 rotates the first rotating shaft 63 about the second rotation axis AX2, and rotates the nozzle arm 61 along a substantially horizontal plane about the first rotating shaft 63. As a result, the first nozzle 41 moves along a substantially horizontal plane. Specifically, the first nozzle 41 revolves around the first rotating shaft 63 about the second rotation axis AX2. The first driving unit 65 includes, for example, a stepping motor.

[0034] The second nozzle 71 supplies the rinse liquid to the substrate W. Specifically, the second nozzle 71 discharges the rinse liquid toward the rotating substrate W. The rinse liquid supply unit 7 supplies the rinse liquid to the second nozzle 71. Specifically, the rinse liquid is supplied to the second nozzle 71 through the second supply pipe 72. The second supply pipe 72 is a tubular member through which the rinse liquid flows. The rinse liquid is, for example, deionized water, carbonated water, electrolyzed ion water, hydrogen water, ozone water, or hydrochloric acid water with a dilution concentration (for example, about 10 ppm to 100 ppm). The second nozzle 71 discharges the rinse liquid in a stationary state. The second nozzle 71 may be referred to as a fixed nozzle. Note that the second nozzle 71 may be a scan nozzle.

[0035] Each of the guards 10 has a substantially cylindrical shape. The plurality of guards 10 receive the etching liquid and the rinse liquid discharged from the substrate W.

[0036] The measurement unit 8 acquires surface information regarding the surface of the substrate W. The surface information includes, for example, information indicating the thickness distribution of the substrate W. Further, the surface information includes, for example, information indicating the surface shape (profile) of the substrate W. Note that based on one of the information on the thickness distribution of the substrate W and the information on the surface shape of the substrate W, the other can be obtained. That is, acquiring the information on the thickness distribution of the substrate W and acquiring the information on the surface shape of the substrate W are substantially the same.

[0037] In this embodiment, the measuring unit 8 measures the thickness of the object TG in a non-contact manner and generates a thickness detection signal indicating the thickness of the object TG. The thickness detection signal is input to the control device 101.

[0038] The measuring unit 8 measures the thickness of the object TG by, for example, the spectral interference method. Specifically, the measuring unit 8 includes an optical probe 81, a signal line 83, and a measuring instrument 85. The optical probe 81 has a lens. The signal line 83 connects the optical probe 81 and the measuring instrument 85. The signal line 83 includes, for example, an optical fiber. The measuring instrument 85 has a light source and a light receiving element. The light emitted from the light source of the measuring instrument 85 is emitted to the object TG via the signal line 83 and the optical probe 81. The light reflected by the object TG is received by the light receiving element of the measuring instrument 85 via the optical probe 81 and the signal line 83. The measuring instrument 85 analyzes the light received by the light receiving element and calculates the thickness of the object TG. The measuring instrument 85 generates a thickness detection signal indicating the calculated thickness of the object TG. Note that the measuring method of the measuring unit 8 is not limited to the spectral interference method, and other measuring methods may be used as long as the thickness of the object TG can be measured.

[0039] The probe moving mechanism 9 moves the optical probe 81 in a substantially horizontal direction. Specifically, the probe moving mechanism 9 pivots the optical probe 81 about a third rotation axis AX3 along a substantially vertical direction. The optical probe 81 emits light toward the substrate W while moving (while pivoting). Therefore, the thickness detection signal indicates the thickness distribution of the object TG.

[0040] Specifically, the probe movement mechanism 9 includes a probe arm 91, a second rotation shaft 93, and a second drive unit 95. The probe arm 91 extends along a substantially horizontal direction. An optical probe 81 is disposed at the tip of the probe arm 91. The probe arm 91 is coupled to the second rotation shaft 93. The second rotation shaft 93 extends along a substantially vertical direction. The second drive unit 95 rotates the second rotation shaft 93 about a third rotation axis AX3, and rotates the probe arm 91 along a substantially horizontal plane about the second rotation shaft 93. As a result, the optical probe 81 moves along a substantially horizontal plane. Specifically, the optical probe 81 orbits around the second rotation shaft 93 about the third rotation axis AX3. The second drive unit 95 includes, for example, a stepping motor.

[0041] In the present embodiment, the measurement unit 8 is used to detect surface information regarding the surface of the substrate W to be processed.

[0042] Also, in the present embodiment, the measurement unit 8 is used to detect the processing amount. The processing amount indicates the amount by which the substrate W is processed when the processing unit 1 processes the substrate W. Specifically, the measurement unit 8 acquires the surface information of the substrate W before processing (hereinafter sometimes referred to as pre-processing surface information) and the surface information of the substrate W after processing (hereinafter sometimes referred to as post-processing surface information). The processing amount can be calculated from the difference between the pre-processing surface information and the post-processing surface information. In the present embodiment, the processing amount indicates the amount by which the substrate W is processed by the etching solution (processing solution) supplied from the first nozzle 41 to the substrate W. In the present embodiment, the processing amount indicates the etching amount. The etching amount indicates the difference between the thickness of the object TG before the etching process and the thickness of the object TG after the etching process.

[0043] The control device 101 acquires the etching amount by calculating the etching amount based on the thickness detection signal input from the measurement unit 8 (measuring instrument 85) during the execution of machine learning. More specifically, the control device 101 acquires the distribution of the etching amount (distribution of the processing amount). The control device 101 generates learning data (learning data set) for machine learning using the acquired etching amount.

[0044] Further, during the manufacture of the semiconductor product, the control device 101 may calculate the etching amount based on the thickness detection signal input from the measurement unit 8 (measuring instrument 85) and acquire the etching amount.

[0045] In addition, a rotation speed signal is input to the control device 101 from the encoder 55. Note that the rotation speed of the substrate W during processing is, for example, constant. Specifically, as will be described with reference to FIG. 9, the control device 101 stores a recipe 131 for controlling each part of the substrate processing apparatus 100, and the recipe 131 indicates a set value of the rotation speed of the motor main body 51. The control device 101 refers to the recipe 131 and controls the processing executed by the processing unit 1.

[0046] Subsequently, with reference to FIG. 3, the scan processing of the substrate W by the first nozzle 41 will be described. FIG. 3 is a plan view showing the scan processing of the present embodiment. As shown in FIG. 3, the scan processing is a process in which the first nozzle 41 moves while discharging the processing liquid onto the object TG so that the liquid landing position of the processing liquid on the surface of the object TG forms an arc-shaped locus TJ1 in a plan view. The locus TJ1 passes through the central portion CT of the substrate W. The central portion CT indicates a portion of the substrate W through which the first rotation axis AX1 passes. The scan processing is executed while the substrate W is rotating.

[0047] In the present embodiment, the first nozzle 41 discharges the etching liquid toward the rotating substrate W while moving from the first position X1 to the ninth position X9. Each position X1 to X9 included from the first position X1 to the ninth position X9 is included in the locus TJ1. The section from the first position X1 to the ninth position X9 indicates the moving section in which the first nozzle 41 moves.

[0048] Among the first position X1 to the ninth position X9, the first position X1 indicates the start position of the discharge of the processing liquid (etching liquid), and the ninth position X9 indicates the stop position of the discharge of the processing liquid (etching liquid). The moving speed of the first nozzle 41 at the first position X1 is 0 mm / s, and the moving speed of the first nozzle 41 at the ninth position X9 is 0 mm / s. Therefore, the first position X1 is the start position of the scan process, and the ninth position X9 is the end position of the scan process. Also, the first position X1 is the start position of the movement of the first nozzle 41, and the ninth position X9 is the end position of the movement of the first nozzle 41. In the following description, the moving speed of the first nozzle 41 during the scan process may be referred to as the "scan speed".

[0049] During the scan process, the first nozzle 41 passes through each intermediate position (each position X2 to X8 from the second position X2 to the eighth position X8) between the first position X1 and the ninth position X9.

[0050] Subsequently, referring to FIG. 4, the scan speed information will be described. The scan speed information indicates the set value of the moving speed of the first nozzle 41 (the set value of the scan speed) during the scan process. FIG. 4 is a diagram showing the scan speed information of the present embodiment. Note that the scan speed is an example of the "speed condition" of the present invention. Specifically, FIG. 4 shows the relationship between each position X1 to X9 included in the moving section of the first nozzle 41 described with reference to FIG. 3 and the set value of the scan speed.

[0051] In FIG. 4, the upper column indicates each position X1 to X9 included in the moving section of the first nozzle 41, and the lower column indicates the set value of the scan speed. Each position X1 to X9 included in the moving section of the first nozzle 41 is defined by the radial position of the substrate W. Specifically, the upper column indicates the start position of the moving section of the first nozzle 41 (the start position of the movement of the first nozzle 41), the end position of the moving section of the first nozzle 41 (the end position of the movement of the first nozzle 41), and a plurality of intermediate positions between the start position and the end position of the moving section of the first nozzle 41 (a plurality of positions through which the first nozzle 41 passes).

[0052] As shown in FIG. 4, the scan speed information indicates the set value of the scan speed for each position X1 to X9 included in the movement section of the first nozzle 41. Hereinafter, each position X1 to X9 included in the movement section of the first nozzle 41 may be referred to as a "speed setting position". In the present embodiment, the scan speed information indicates nine speed setting positions.

[0053] Specifically, each speed setting position corresponds to each position X1 to X9 from the first position X1 to the ninth position X9 described with reference to FIG. 3. As described with reference to FIG. 3, the scan speed set at the start position (first position X1) of the movement section of the first nozzle 41 is 0 [mm / s], and the scan speed set at the end position (ninth position X9) of the movement section of the first nozzle 41 is 0 [mm / s].

[0054] The control device 101 described with reference to FIGS. 1 and 2 controls the nozzle movement mechanism 6 (first drive section 65) with reference to FIG. 2 based on the scan speed information. As a result, the first nozzle 41 moves along the locus TJ1 described with reference to FIG. 3 such that the scan speed at each speed setting position becomes the scan speed defined by the scan speed information.

[0055] Subsequently, with reference to FIG. 5, the movement speed (scan speed) of the first nozzle 41 during the scan process will be described. FIG. 5 is a graph showing an example of the movement speed of the first nozzle 41 in the present embodiment.

[0056] In FIG. 5, the vertical axis represents the scan speed [mm / s], and the horizontal axis represents the radial position [mm] of the substrate W. As shown in FIG. 5, the scan speed at the start of the scan process is 0 [mm / s]. Also, the scan speed at the end of the scan process is 0 [mm / s].

[0057] As described with reference to FIGS. 3 and 4, the scan speed is set for each of the positions X1 to X9 (each speed setting position) from the first position X1 to the ninth position X9. As a result, as shown in FIG. 5, between adjacent speed setting positions, the scan speed continuously changes from the scan speed set for one speed setting position to the scan speed set for the other speed setting position. For example, as shown in FIG. 4, a scan speed Y3 is set for the third position X3, and a scan speed Y4 is set for the fourth position X4. Therefore, the scan speed of the first nozzle 41 continuously changes from the scan speed Y3 to the scan speed Y4 while the first nozzle 41 moves from the third position X3 to the fourth position X4.

[0058] Next, with reference to FIG. 6, the thickness measurement process by the measurement unit 8 will be described. FIG. 6 is a plan view showing the thickness measurement process of the present embodiment. As shown in FIG. 6, the thickness measurement process is a process of measuring the thickness of the object TG while the optical probe 81 moves so that the measurement positions of the thickness with respect to the object TG form an arc-shaped locus TJ2 in a plan view. The locus TJ2 passes through the edge portion EG of the substrate W and the central portion CT of the substrate W. The edge portion EG indicates the peripheral portion of the substrate W. The thickness measurement process is executed while the substrate W is rotating.

[0059] Specifically, the optical probe 81 emits light toward the object TG while moving between the central portion CT and the edge portion EG of the substrate W in a plan view. As a result, the thickness of the object TG is measured at each measurement position included in the locus TJ2. Each measurement position corresponds to each radial position of the substrate W. Therefore, by the thickness measurement process, the thickness distribution of the object TG in the radial direction RD of the substrate W is measured. Note that the surface shape (profile) of the object TG coincides with the shape indicating the thickness distribution of the object TG.

[0060] Next, referring to FIG. 7, the etching liquid supply unit 4 of the present embodiment will be described. FIG. 7 is a schematic diagram of the etching liquid supply unit 4 of the present embodiment. As shown in FIG. 7, in addition to the first supply pipe 42 described with reference to FIG. 2, the etching liquid supply unit 4 further includes a temperature sensor 421, a concentration sensor 422, a valve 423, a mixing valve 424, a flow meter 425, and a heater 426.

[0061] The temperature sensor 421 measures the temperature of the etching liquid flowing through the first supply pipe 42. The temperature sensor 421 generates a temperature signal indicating the temperature of the etching liquid and inputs it to the control device 101. The temperature of the etching liquid during processing is, for example, constant.

[0062] The concentration sensor 422 measures the concentration of the etching component contained in the etching liquid flowing through the first supply pipe 42. The concentration sensor 422 generates a concentration signal indicating the concentration of the etching liquid and inputs it to the control device 101. The concentration of the etching liquid during processing is, for example, constant.

[0063] The valve 423 is disposed in the first supply pipe 42. The valve 423 switches the supply and supply stop of the etching liquid to the first nozzle 41. Specifically, when the valve 423 opens, the etching liquid is discharged from the first nozzle 41 toward the substrate W. On the other hand, when the valve 423 closes, the discharge of the etching liquid stops. Further, the valve 423 controls the flow rate of the etching liquid flowing downstream of the valve 423 in the first supply pipe 42. Specifically, the flow rate of the etching liquid flowing downstream of the valve 423 is adjusted according to the opening degree of the valve 423. Therefore, the discharge flow rate of the etching liquid is adjusted according to the opening degree of the valve 423. The valve 423 is, for example, a motor valve.

[0064] The mixing valve 424 is disposed in the first supply pipe 42. When the mixing valve 424 opens, pure water flows into the first supply pipe 42 and the concentration of the etching liquid is diluted.

[0065] The flow meter 425 measures the discharge flow rate of the etching solution. Specifically, the object to be measured by the flow meter 425 is the flow rate of the etching solution flowing through the first supply pipe 42. The flow meter 425 generates a discharge flow rate signal indicating the discharge flow rate of the etching solution and inputs it to the control device 101. The discharge flow rate of the etching solution during processing is, for example, constant.

[0066] The heater 426 heats the etching solution flowing through the first supply pipe 42.

[0067] Next, with reference to FIGS. 8A to 8C, the etching process for the substrate W (specifically, the object TG) by the substrate processing apparatus 100 will be described. FIG. 8A is a diagram showing the thickness patterns TA, TB, and TC before etching. FIG. 8B is a diagram showing the etching rate distributions EA, EB, and EC. FIG. 8C is a diagram showing the thickness patterns TAR, TBR, and TCR after etching. The "thickness pattern" indicates the distribution of a plurality of thicknesses measured at a plurality of measurement positions of the object TG. That is, the "thickness pattern" indicates the distribution of the thickness of the object TG in the radial direction RD of the substrate W.

[0068] In FIGS. 8A to 8C, the horizontal axis indicates the position on the substrate W. In FIGS. 8A and 8C, the vertical axis indicates the thickness of the object TG. In FIG. 8B, the vertical axis indicates the etching rate of the object TG. The etching rate is the amount of etching per unit time.

[0069] As shown in FIG. 8A, when the object TG has a thickness pattern TA before etching, as shown in FIG. 8B, the speed conditions are determined so as to have an etching rate distribution EA highly correlated with the thickness pattern TA, and the etching process is executed. As a result, according to the present embodiment, as shown in FIG. 8C, when the object TG has a thickness pattern TA before etching, the thickness pattern TAR of the object TG after etching becomes substantially flat. That is, in a wide area of the object TG (the area from the central portion CT to the edge portion EG), the variation in the thickness of the object TG after etching can be suppressed.

[0070] Similarly, as shown in FIGS. 8A and 8B, the etching process is executed with the speed condition determined such that the etching rate distribution EB highly correlated with the thickness pattern TB of the object TG before etching is obtained. As a result, as shown in FIG. 8C, when the object TG has the thickness pattern TB before etching, the thickness pattern TBR of the object TG after etching becomes substantially flat.

[0071] Similarly, as shown in FIGS. 8A and 8B, the etching process is executed with the speed condition determined such that the etching rate distribution EC highly correlated with the thickness pattern TC of the object TG before etching is obtained. As a result, as shown in FIG. 8C, when the object TG has the thickness pattern TC before etching, the thickness pattern TCR of the object TG after etching becomes substantially flat.

[0072] Next, the control device 101 will be described with reference to FIG. 9. FIG. 9 is a block diagram of the control device 101 of the present embodiment. As shown in FIG. 9, the control device 101 includes a control unit 102, a storage unit 103, an input unit 104, and a display unit 105.

[0073] The control unit 102 includes a processor. The control unit 102 includes, for example, a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). Alternatively, the control unit 102 may include a general-purpose arithmetic unit or a dedicated arithmetic unit. The control unit 102 may further include an NPU (Neural Network Processing Unit).

[0074] The storage unit 103 stores data and computer programs. The storage unit 103 has a main storage device. The main storage device is, for example, a semiconductor memory. The storage unit 103 may further have an auxiliary storage device. The auxiliary storage device is, for example, a semiconductor memory and / or a hard disk drive. The storage unit 103 may have a removable medium. The control unit 102 controls the operations of each part of the substrate processing apparatus 100 based on the data and computer programs stored in the storage unit 103. Further, the control unit 102 executes machine learning based on the data and computer programs stored in the storage unit 103.

[0075] Specifically, the storage unit 103 stores a recipe 131, a control program 132, a learned model 133, and a learning program 134. The recipe 131 defines the processing content and processing procedure of the substrate W. Also, the recipe 131 indicates various setting values.

[0076] The control unit 102 controls the operations of each part of the substrate processing apparatus 100 based on the recipe 131 and the control program 132. The learned model 133 outputs speed conditions based on input data. In the present embodiment, the learned model 133 outputs information indicating the scan speed based on the surface information as the input data. The learning program 134 is a program for executing an algorithm that finds a certain rule from the learning data set and generates a model (learned model 133) that represents the rule.

[0077] The storage unit 103 stores learning target speed information ♯A1 to ♯An when machine learning is executed. Here, "n" represents a positive integer. Each of the learning target speed information ♯A1 to ♯An indicates scan speed information, similar to the scan speed information described with reference to FIG. 4. The content of the set of scan speeds set for each speed setting position described with reference to FIG. 4 is different for the learning target speed information ♯A1 to ♯An.

[0078] The input unit 104 receives an input from an operator and outputs information indicating the input result to the control unit 102. For example, the input unit 104 receives an input of learning target speed information #A1 to #An. The input unit 104 includes, for example, a touch panel and a pointing device. The touch panel is disposed on the display surface of the display unit 105, for example. The input unit 104 and the display unit 105 constitute a graphical user interface, for example.

[0079] The display unit 105 displays various information. In the present embodiment, the display unit 105 displays, for example, various error screens and various setting screens (input screens). The display unit 105 has, for example, a liquid crystal display or an organic EL (electroluminescence) display.

[0080] Note that the control device 101 may have an interface for acquiring the learning target speed information #A1 to #An from a removable memory. The removable memory includes, for example, a USB (Universal Serial Bus) memory. Alternatively, the control device 101 may have a drive for acquiring the learning target speed information #A1 to #An from a removable medium. The removable medium includes, for example, an optical disk such as a CD (Compact Disc).

[0081] Subsequently, with reference to FIGS. 1, 2, 9, and 10, a method for generating learning data executed by the substrate processing apparatus 100 will be described. FIG. 10 is a flowchart showing the method for generating learning data in the present embodiment. Specifically, FIG. 10 shows the processing executed by the control unit 102 during the generation of learning data. The method for generating learning data in the present embodiment includes each process of steps S1 to S9.

[0082] The process shown in FIG. 10 is started when the operator operates the input unit 104. During the generation of learning data, a plurality of substrates W to be learned are accommodated in at least one of the plurality of load ports LP. Note that the substrate W to be learned is the same type of substrate W as the substrate W used in the manufacture of semiconductor products.

[0083] When generating learning data, the control unit 102 first selects one of the plurality of processing units 1 as the processing unit 1 to be learned. Note that the processes described with reference to FIG. 10 are sequentially executed for the plurality of processing units 1 included in the substrate processing apparatus 100.

[0084] When the control unit 102 determines the processing unit 1 to be learned, it selects one of the learning target speed information #A1 to #An and acquires one of the learning target speed information #A1 to #An (step S1). Hereinafter, one of the learning target speed information #A1 to #An is referred to as "learning target speed information #A". For example, the control unit 102 selects the learning target speed information #A in ascending order of the identification numbers assigned to the learning target speed information #A1 to #An.

[0085] After the control unit 102 selects one of the learning target speed information #A1 to #An, it controls the indexer robot IR and the center robot CR so that the substrate W to be learned is carried into the chamber 2 of the processing unit 1 to be learned (step S2). The control unit 102 holds the substrate W to be learned carried into the chamber 2 by the spin chuck 3.

[0086] When the substrate W to be learned is held by the spin chuck 3, the control unit 102 causes the measuring unit 8 to measure the thickness distribution of the object TG included in the substrate W to be learned (step S3). The thickness distribution of the object TG measured here indicates the thickness distribution of the object TG before the etching process. Hereinafter, the thickness distribution of the object TG before the etching process may be referred to as "the thickness distribution before processing".

[0087] After measuring the thickness distribution before processing, the control unit 102 controls the operations of each part of the substrate processing apparatus 100 so that an etching process is executed on the substrate W to be learned (step S4). Specifically, as described with reference to FIGS. 3 to 5, while the first nozzle 41 moves (swivels) at a speed based on the learning target speed information #A, the control unit 102 controls the operations of each part of the substrate processing apparatus 100 so that an etching solution is supplied from the first nozzle 41 to the substrate W to be learned, and executes an etching process on the substrate W to be learned.

[0088] After executing the etching process, the control unit 102 causes the measuring unit 8 to measure the thickness distribution of the object TG included in the substrate W to be learned (step S5). The thickness distribution of the object TG measured here indicates the thickness distribution of the object TG after the etching process. Hereinafter, the thickness distribution of the object TG after the etching process may be referred to as the "thickness distribution after processing". As described with reference to FIG. 11, the thickness distribution after processing indicates the thickness distribution after the drying process.

[0089] After measuring the thickness distribution after processing, the control unit 102 releases the holding of the substrate W to be learned by the spin chuck 3, and causes the indexer robot IR to carry out the substrate W to be learned from the chamber 2 (step S6). Then, the control unit 102 controls the indexer robot IR and the center robot CR so that the substrate W to be learned is transported to one of the plurality of load ports LP.

[0090] Next, the control unit 102 acquires throughput information indicating the throughput (etching amount) (step S7). Specifically, the control unit 102 calculates the difference between the thickness distribution before processing and the thickness distribution after processing, and acquires the throughput information.

[0091] When the control unit 102 acquires the throughput (etching amount) information, it generates learning data based on the learning target speed information #A and the throughput information (step S8). The learning data indicates the learning target speed information #A and the throughput information. The learning data is stored in the storage unit 103. Then, the process shown in FIG. 10 ends.

[0092] The control unit 102 repeatedly executes the process shown in FIG. 10 until all of the learning target speed information #A1 to #An is selected. As a result, a plurality of learning data is stored in the storage unit 103. In the present embodiment, as will be described later, a plurality of learning data sets are generated based on this learning data. Further, at least one of the processes in step S7 and step S8 may be executed between the process in step S5 and the process in step S6.

[0093] Subsequently, with reference to FIGS. 2, 9, and 11, the etching process (step S4) will be described. FIG. 11 is a flowchart showing the etching process of the present embodiment.

[0094] As shown in FIG. 11, after the spin chuck 3 holds the substrate W to be learned, the substrate W to be learned is processed (step S41). Specifically, the control unit 102 rotates the substrate W held by the spin chuck 3 in the spin motor unit 5. Thereafter, the control unit 102 controls the nozzle moving mechanism 6 and the etching solution supply unit 4 so that the etching solution is supplied from the first nozzle 41 toward the substrate W to be learned, as described with reference to FIG. 10. As a result, the substrate W to be learned is etched.

[0095] When the etching of the substrate W to be learned is completed, the control unit 102 controls the rinse solution supply unit 7 to supply the rinse solution to the substrate W to be learned, thereby removing the etching solution from the substrate W to be learned (step S42). Specifically, the etching solution is flushed outward of the substrate W to be learned by the rinse solution and discharged around the substrate W to be learned. As a result, the liquid film of the etching solution on the substrate W to be learned is replaced with the liquid film of the rinse solution.

[0096] After replacing the etching solution with the rinse solution, the control unit 102 controls the spin motor unit 5 to dry the substrate W to be learned (step S43). As a result, the process shown in FIG. 11 ends. Specifically, the control unit 102 increases the rotation speed of the substrate W to be learned to be higher than the rotation speeds during the etching process and the rinse process. As a result, a large centrifugal force is applied to the rinse solution on the substrate W to be learned, and the rinse solution adhering to the substrate W to be learned is shaken off around the substrate W to be learned. In this way, the rinse solution is removed from the substrate W to be learned, and the substrate W to be learned is dried. Note that the control unit 102 stops the rotation of the substrate W by the spin motor unit 5 after a predetermined time has elapsed since the start of the high-speed rotation of the substrate W to be learned, for example.

[0097] Next, with reference to FIGS. 12 to 14, a method for generating the learned model 133 based on a plurality of learning data will be described. In the present embodiment, a plurality of learning data sets are generated based on a plurality of learning data, and a learned model 133 is generated for each of the plurality of learning data sets. That is, a plurality of learned models 133 are generated based on a plurality of learning data. Hereinafter, a specific description will be given.

[0098] FIG. 12 is a flowchart showing a method for generating the learned model 133 of the present embodiment. The method for generating the learned model 133 of the present embodiment includes steps S101 to S104. FIG. 13 is a diagram showing the learning data table TB10 of the present embodiment. FIG. 14 is an image diagram showing the result of clustering a plurality of throughput information of the present embodiment into a plurality of clusters.

[0099] As shown in FIG. 12, in step S101, the control unit 102 clusters a plurality of throughput (etching amount) information stored in the storage unit 103 into a plurality of clusters. That is, the control unit 102 clusters the throughput information when processing a plurality of substrates W to be learned under a plurality of speed conditions (scan speeds) into a plurality of clusters.

[0100] Specifically, as shown in FIG. 13, the learning data table TB10 is stored in the storage unit 103. The learning data table TB10 includes learning target speed information ♯A, which is scan speed information, and processing amount (etching amount) information in a state where they are associated with each other. The control unit 102 clusters a plurality (for example, several hundred) of processing amount information included in the learning data table TB10 into a plurality (for example, several to several tens) of clusters. Clustering is to find information with similarity or correlation and group the information with similarity or correlation. Therefore, by clustering, information with similarity or correlation is classified into one cluster.

[0101] By the clustering process, for example, a cluster data table TB20 as shown in FIG. 14 is obtained. That is, data with similar etching profiles, which are processing amount (etching amount) information, are grouped together. As shown in FIG. 14, each cluster includes one or more, for example, several to several hundred or less, of processing amount information. In FIG. 14, the processing amount information is shown in a graph.

[0102] Also, the clustering method is not particularly limited. For example, machine learning can be used. Also, as the clustering method, unsupervised learning may be used, for example, the k-means method may be used. Also, the number of clusters is not particularly limited as long as it is two or more. For example, the optimal number may be determined using the elbow method or the like. Note that, as the clustering method, a method other than machine learning may be used. For example, grouping rules may be set and clustering may be performed according to the rules, or the user may perform clustering by judging from the shape of the graph.

[0103] Next, in step S102, the control unit 102 selects a cluster to be used for the learning dataset from among the plurality of clusters. Specifically, the control unit 102 selects a cluster that includes a predetermined number (threshold value) or more substrates W. That is, for clusters in which the number of substrates W is less than the predetermined number (threshold value), they are not used as the learning dataset. The predetermined number (threshold value) is not particularly limited, but for example, it is 5, 10, or 20.

[0104] By the cluster selection process, for example, cluster CL1, cluster CL2, and cluster CL3 in FIG. 14 are selected. On the other hand, cluster CL4 is not selected. That is, for cluster CL4, since the throughput information is small, it is not used as the learning dataset.

[0105] Next, in step S103, the control unit 102 selects the throughput information to be used for the learning data in each cluster. In the present embodiment, the control unit 102 selects throughput information in each cluster that has a correlation with other throughput information equal to or higher than a predetermined condition. That is, for throughput information with a relatively low correlation with other throughput information, it is not used as the learning dataset.

[0106] Specifically, by the selection of the throughput information to be used for the learning data, for example, in cluster CL3 of FIG. 14, throughput information excluding throughput information PI with a relatively low correlation with other throughput information is selected. That is, among the plurality of throughput information in cluster CL3, throughput information PI with a relatively low correlation with other throughput information is not used as the learning dataset.

[0107] The method for deriving the correlation with other throughput information is not particularly limited, but statistical analysis can be used, and for example, RMSE (root mean square error) may be used. That is, reference throughput information may be selected from the throughput information of each cluster, and throughput information with a correlation with the reference throughput information equal to or higher than a predetermined condition may be selected.

[0108] As described above, a plurality of clusters and a plurality of throughput information used for generating the learned model 133 are selected. Then, based on the plurality of throughput information used for generating the learned model 133 and the learning target speed information #A corresponding to this throughput information, a plurality of learning datasets used for generating the learned model 133 are generated. Note that the number of learning datasets is equal to the number of clusters selected in step S102. Also, here, after selecting the clusters to be used for the learning dataset from the plurality of clusters (step S102), an example is shown in which the throughput information to be used for the learning data is selected for each cluster (step S103). However, after selecting the throughput information to be used for the learning data for each cluster, the clusters to be used for the learning dataset may be selected from the plurality of clusters.

[0109] Next, in step S104, the control unit 102 generates a plurality of learned models 133. Specifically, the control unit 102 generates the learned model 133 by performing machine learning on the learning dataset (a plurality of learning data) based on the learning program 134. Here, learning means discovering a certain rule from the learning dataset. In the present embodiment, the learning data shows the relationship between the scan speed information (the scan speed at each speed setting position) and the throughput (etching amount) information. The control unit 102 learns the learning dataset to discover a certain rule between the scan speed information and the throughput information.

[0110] In the present embodiment, the "input" is the surface information before processing (hereinafter, may be described as the surface information before processing), and the "output" is the scan speed information. In other words, the surface information before processing is the explanatory variable, and the scan speed (speed condition) information output from the learned model 133 is the target variable. Note that since the target value of the throughput (etching amount) is derived based on the surface information before processing, inputting the surface information before processing and inputting the target value of the throughput (etching amount) are substantially the same.

[0111] In addition, the machine learning algorithm for constructing the learned model 133 is not particularly limited as long as it is supervised learning. For example, it can be a decision tree, a nearest neighbor method, a simple Bayesian classifier, a support vector machine, or a neural network. Therefore, the learned model 133 includes a decision tree, a nearest neighbor method, a simple Bayesian classifier, a support vector machine, or a neural network. The error backpropagation method may be used for machine learning.

[0112] As described above with reference to FIGS. 12 to 14, in this embodiment, the plurality of learned models 133 are obtained by clustering the throughput (etching amount) information when processing a plurality of substrates W to be learned under a plurality of speed conditions (scan speeds) into a plurality of clusters and performing machine learning using the speed conditions and throughput information of each cluster. Therefore, the learned models 133 are generated for each throughput information having similarity or correlation. Thus, for example, the prediction accuracy of the learned model 133 can be improved as compared with the case of generating one learned model 133 using all the throughput information.

[0113] In addition, in this embodiment, a cluster having a predetermined number (threshold) or more of substrates W to be learned is selected from the plurality of clusters, and a plurality of learned models 133 are obtained by performing machine learning using the speed conditions and throughput information of the selected clusters. That is, for a cluster having a predetermined number or more of learning data (speed conditions and throughput information), a learned model 133 is generated. On the other hand, for a cluster having less learning data (speed conditions and throughput information), a learned model 133 is not generated. As a result, it is possible to suppress the generation of a learned model 133 with low prediction accuracy.

[0114] Also, in the present embodiment, in each cluster, throughput information having a correlation with other throughput information higher than a predetermined condition is selected, and the learned model 133 is obtained by performing machine learning using the selected throughput information. That is, the learned model 133 is generated using throughput information having a correlation with other throughput information higher than a predetermined condition. On the other hand, throughput information having a low correlation with other throughput information is not used for generating the learned model 133. As a result, it is possible to suppress a decrease in the prediction accuracy of the learned model 133.

[0115] Subsequently, with reference to FIGS. 1, 2, and 15, a substrate processing method executed by the substrate processing apparatus 100 will be described. FIG. 15 is a flowchart showing the substrate processing method in the present embodiment. Specifically, FIG. 15 shows the processing executed by the control unit 102 when etching the substrate W to be processed. The substrate processing method of the present embodiment includes each process from step S201 to step S206.

[0116] The process shown in FIG. 15 is started by an operator operating the input unit 104. A plurality of substrates W to be processed are accommodated in at least one of the plurality of load ports LP.

[0117] When etching the substrate W to be processed, the control unit 102 controls the indexer robot IR and the center robot CR so that the substrate W to be processed is carried into the chamber 2 of one of the plurality of processing units 1 (step S201). The control unit 102 holds the substrate W to be processed on the spin chuck 3.

[0118] When the substrate W to be processed is held on the spin chuck 3, the control unit 102 causes the measuring unit 8 to measure the thickness distribution of the object TG included in the substrate W to be processed (step S202). In other words, the thickness distribution before processing is measured. More specifically, the control unit 102 acquires pre-processing surface information regarding the surface of the substrate W to be processed.

[0119] Next, the control unit 102 selects one learned model 133 from the plurality of learned models 133 based on the thickness distribution of the object TG (step S203). In other words, the control unit 102 selects one learned model 133 from the plurality of learned models 133 based on the pre-processing surface information.

[0120] Specifically, as described with reference to FIGS. 8A to 8C, the control unit 102 selects the learned model 133 having the etching rate distribution with the highest correlation with the thickness pattern of the object TG. In other words, the control unit 102 selects the learned model 133 among the plurality of learned models 133 whose processing amount information shape (see FIG. 14) is closest to the pre-processing surface information (profile) of the substrate W to be processed. The control unit 102 may select one learned model 133 from the plurality of learned models 133 using machine learning such as the k-nearest neighbor method.

[0121] Furthermore, in other words, in step S203, the control unit 102 selects one learned model 133 from the plurality of learned models 133 based on the target value of the processing amount. Specifically, the control unit 102 derives the target value of the processing amount based on the pre-processing surface information. Then, the control unit 102 selects one learned model 133 obtained using the processing amount information of one cluster having a correlation with the target value of the processing amount equal to or higher than a predetermined condition from the plurality of learned models 133. That is, the control unit 102 selects one cluster having the highest correlation between the plurality of processing amount information included in each cluster and the target value of the processing amount. Then, the control unit 102 selects one learned model 133 corresponding to the selected cluster. That is, the control unit 102 selects one learned model 133 obtained using the processing amount information of the selected cluster.

[0122] Next, the control unit 102 inputs the pre-processing surface information to the learned model 133 and outputs the scan speed (speed condition) information from the learned model 133, thereby obtaining the scan speed (speed condition) information, which is the target variable (step S204).

[0123] When the scan speed (speed condition) information is acquired, the control unit 102 controls the operations of each part of the substrate processing apparatus 100 so that an etching process is executed on the substrate W to be processed (step S205). Specifically, as described with reference to FIGS. 3 to 5, while the first nozzle 41 moves (swivels) at a speed based on the scan speed information, the control unit 102 controls the operations of each part of the substrate processing apparatus 100 so that an etching liquid is supplied from the first nozzle 41 to the substrate W to be processed, and executes an etching process on the substrate W to be processed. More specifically, the substrate W to be processed is etched by the same process as the process described with reference to FIG. 11.

[0124] Next, the control unit 102 releases the holding of the substrate W to be processed by the spin chuck 3, and causes the indexer robot IR to carry out the substrate W to be processed from the chamber 2 (step S206). Thereafter, the control unit 102 controls the indexer robot IR and the center robot CR so that the substrate W to be processed is conveyed to one of the plurality of load ports LP.

[0125] As described above, the processing of the substrate W to be processed is completed.

[0126] As described above, an embodiment of the present invention has been described with reference to FIGS. 1 to 15. In this embodiment, as described above, pre-process surface information (distribution of the thickness of the object TG) regarding the surface of the substrate W to be processed is acquired, and one learned model 133 is selected from a plurality of learned models 133 based on the pre-process surface information. Then, the substrate W to be processed is processed at a speed condition (scan speed) obtained based on the pre-process surface information and the selected learned model 133. Therefore, the thickness of the object TG constituting the substrate W can be made a desired thickness.

[0127] Incidentally, although details are omitted, the inventor of the present application generated one learned model using all the data in the learning data table TB10 shown in FIG. 13 and the substrate W (hereinafter referred to as substrate Wa) processed by the substrate processing method of the present embodiment, and performed an experiment to compare the variation in the thickness of the object TG between the substrate W (hereinafter referred to as substrate Wb) processed using the learned model. In this experiment, it was confirmed that the variation in the thickness of the object TG of substrate Wa is smaller than the variation in the thickness of the object TG of substrate Wb.

[0128] Also, in the present embodiment, based on the pre-processing surface information, a target value of the processing amount is derived. Then, one learned model 133 is selected from the plurality of learned models 133 using the processing amount information of one cluster having a correlation with the target value of the processing amount equal to or higher than a predetermined condition. Therefore, the thickness of the object TG can be easily made the desired thickness. For example, the thickness pattern of the object TG can also be made the desired pattern.

[0129] Also, in the present embodiment, the pre-processing surface information includes information indicating the surface shape (thickness distribution) of the substrate W to be processed. Therefore, one learned model 133 can be selected from the plurality of learned models 133 based on the surface shape (thickness distribution) of the substrate W to be processed. Further, the substrate W can be processed under the speed condition obtained based on the surface shape (thickness distribution) of the substrate W to be processed and the selected learned model 133. As a result, the variation in the thickness of the object TG constituting the substrate W can be easily suppressed.

[0130] The embodiments of the present invention have been described above with reference to the drawings. However, the present invention is not limited to the above embodiments, and can be implemented in various aspects without departing from the gist thereof. Also, the plurality of components disclosed in the above embodiments can be modified as appropriate. For example, a certain component among all the components shown in one embodiment may be added to the components of another embodiment, or some of the components among all the components shown in one embodiment may be deleted from the embodiment.

[0131] The drawings schematically show each component mainly for facilitating the understanding of the invention. The thickness, length, number, interval, etc. of each illustrated component may be different from the actual ones for the convenience of drawing preparation. Also, it goes without saying that the configuration of each component shown in the above embodiments is an example and is not particularly limited, and various changes can be made without substantially departing from the effects of the present invention.

[0132] For example, in the embodiments described with reference to FIGS. 1 to 15, the substrate W was a semiconductor wafer, but the substrate W is not limited to a semiconductor wafer. For example, the substrate W can be a substrate for a liquid crystal display device, a substrate for a field emission display (FED), a substrate for an optical disk, a substrate for a magnetic disk, a substrate for a magneto-optical disk, a substrate for a photomask, a ceramic substrate, or a substrate for a solar cell.

[0133] Also, in the above embodiments, the scan speed information defined the moving speed of the first nozzle 41, but the scan speed information may define the relative moving speed between the first nozzle 41 and the substrate W. For example, when the nozzle is fixed and the substrate is movable, the scan speed information may define the moving speed of the substrate with respect to the nozzle. In addition, when the substrate W rotates, the scan speed information defines the relative moving speed between the center of the rotating substrate W and the first nozzle 41.

[0134] Also, in the above embodiments, the first nozzle 41 rotated, but the first nozzle 41 may move linearly.

[0135] Also, in the above embodiments, the first nozzle 41 was a scan nozzle, but the first nozzle 41 may be a fixed nozzle. In this case, the processing unit 1 includes a substrate moving mechanism for moving the substrate W instead of the nozzle moving mechanism 6.

[0136] In the above-described embodiment, an example in which the learned model 133 is stored in the storage unit 103 of the substrate processing apparatus 100 has been shown. However, the present invention is not limited to this. For example, the learned model 133 may be stored in a device different from the substrate processing apparatus 100. That is, the surface information before processing may be transmitted from the substrate processing apparatus 100 to a different device or the like, and the speed condition may be transmitted from the different device or the like to the substrate processing apparatus 100, and the substrate processing apparatus 100 may execute the processing based on this speed condition. In this case, a substrate processing system including the substrate processing apparatus 100 and a device having a storage unit that stores the learned model 133 can also solve the problems of the present application.

[0137] In the above-described embodiment, an example in which the control device 101 of the substrate processing apparatus 100 performs machine learning has been shown. However, the present invention is not limited to this. For example, a learning device different from the substrate processing apparatus 100 may perform machine learning to generate a learned model.

[0138] In the above-described embodiment, an example in which the surface shape of the object TG is made substantially flat by the processing has been shown. However, the present invention is not limited to this. By the processing, the surface shape of the object TG can be made into a desired shape such as a convex shape or a concave shape, for example.

[0139] In the above-described embodiment, the processing executed by the substrate processing apparatus 100 was an etching process. However, the processing executed by the substrate processing apparatus 100 is not limited to the etching process. For example, the processing may be a film forming process.

Industrial Applicability

[0140] The present invention is useful in the field of processing substrates.

Explanation of Signs

[0141] 8: Measurement unit (acquisition unit) 41: First nozzle (nozzle) 100: Substrate processing apparatus 102: Control unit 133: Learned model CL1~CL4: Cluster W: Substrate

Claims

1. A substrate processing method for performing an etching process on a substrate to be processed by supplying a processing liquid from a nozzle to the substrate to be processed, comprising: a step of acquiring pre-processing surface information regarding the surface of the substrate to be processed; a step of selecting one learned model from a plurality of learned models based on the pre-processing surface information; a step of etching the substrate to be processed under a speed condition obtained based on the pre-processing surface information and the selected one learned model; wherein the one learned model outputs the speed condition, which is the moving speed of the nozzle, when the pre-processing surface information is input; the plurality of learned models are a plurality of models obtained by clustering processing amount information regarding the processing amount when processing a plurality of substrates to be learned under a plurality of speed conditions into a plurality of clusters, and performing machine learning using the speed condition and the processing amount information of each cluster; the step of selecting the one learned model derives a target value of the processing amount for etching the substrate to be processed based on the pre-processing surface information; and selects the one learned model obtained using the processing amount information of one cluster having a correlation with the target value of the processing amount higher than a predetermined condition from the plurality of learned models.

2. The substrate processing method according to claim 1, wherein the plurality of learned models are obtained by selecting clusters having a predetermined number or more of the substrates to be learned from the plurality of clusters, and performing machine learning using the speed condition and the processing amount information of the selected clusters.

3. The substrate processing method according to claim 1 or claim 2, wherein the one learned model is obtained by selecting processing amount information having a correlation with other processing amount information higher than a predetermined condition in each cluster, and performing machine learning using the selected processing amount information.

4. The substrate processing method according to any one of claims 1 to 3, wherein the pre-processing surface information includes information indicating the distribution of the thickness of the substrate to be processed.

5. A substrate processing apparatus for performing an etching process on a substrate to be processed by supplying a processing liquid to the substrate, comprising: a nozzle for supplying the processing liquid to the substrate; an acquisition unit for acquiring pre-processing surface information regarding the surface of the substrate to be processed; and a control unit wherein the control unit selects one learned model from a plurality of learned models based on the pre-processing surface information; Etch the substrate to be processed under the speed conditions obtained based on the pre-processing surface information and the selected learned model. When the pre-processing surface information is input into the one learned model, the learned model outputs the speed condition, which is the moving speed of the nozzle. The plurality of learned models are a plurality of models obtained by clustering processing amount information regarding the processing amounts when a plurality of substrates to be learned are processed under a plurality of speed conditions into a plurality of clusters, and performing machine learning using the speed conditions and the processing amount information of each cluster. In the selection of the one learned model, the control unit Derives a target value of the processing amount for etching the substrate to be processed based on the pre-processing surface information. A substrate processing apparatus that selects the one learned model obtained using the processing amount information of one cluster having a correlation with the target value of the processing amount equal to or higher than a predetermined condition from the plurality of learned models.

Images