Double-side polishing apparatus and double-side polishing method

Abstract

To provide a double-side polishing apparatus capable of preventing of a workpiece shape deviating from a target shape at the end of final polishing and obtaining a desired workpiece shape.SOLUTION: A control unit 30 for controlling a polishing machine 10 on the basis of a measurement result of a thickness measuring device 20 includes: a first learner 31 that is trained regarding a correlative relation among main polishing conditions, outer peripheral flatness, and in-plane flatness; a second learner 32 that is trained regarding a correlative relation among end point polishing conditions, degree of change in in-plane flatness when main polishing is stopped, degree of change in outer peripheral flatness when the main polishing is stopped, and an amount of change of outer peripheral flatness during end point polishing; a main polishing condition correction unit 33 that corrects the main polishing conditions at the time of correction under a correction target value of in-plane flatness obtained from learning results of the first learner 31 and main polishing conditions obtained by inputting a target value of outer peripheral flatness to the first learner 31; and a main polishing stopping condition setting unit 34 that sets main polishing stopping conditions using the amount of change in outer peripheral flatness obtained from learning results of the second learner 32.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a double-sided polishing apparatus and a double-sided polishing method for polishing the front and back surfaces of a disk-shaped workpiece.

Background Art

[0002] Conventionally, when polishing the front and back surfaces of a disk-shaped workpiece such as a silicon wafer with upper and lower surface plates, an artificial intelligence model that has learned the relationship between the distance between the surface plates and the flatness of the workpiece is used. For example, an optimal distance between the surface plates obtained by inputting a desired workpiece flatness is used for control. There is known a double-sided polishing apparatus (see, for example, Patent Document 1). Further, there is known a double-sided polishing apparatus that stops double-sided polishing of a workpiece at a timing when the shape index of the entire workpiece becomes a set value of the shape index of the entire workpiece (see, for example, Patent Document 2).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, in order to polish a workpiece in a double-sided polishing apparatus, first, the load applied to the workpiece is gradually increased. When conditions such as the load reaching a predetermined load are satisfied, main polishing is performed in which the workpiece is polished for a predetermined time while maintaining the load applied to the workpiece at the target load. Then, when a predetermined time has elapsed since the start of the main polishing, the main polishing is stopped. After that, the load applied to the workpiece is gradually reduced, and the polishing of the workpiece is completed. Here, even when the load applied to the workpiece after the main polishing is gradually reduced, a change in the shape of the workpiece occurs, and moreover, the amount of the shape change is not constant because it varies depending on polishing conditions, the state of auxiliary materials, etc. Therefore, the operator of the polishing apparatus estimates the amount of shape change of the workpiece after the main polishing, and manually adjusts the timing of stopping the main polishing by the operator. However, since the timing of stopping the main polishing and the like are adjusted depending on the operator's sense, there is a problem that variations occur in the finished quality of the workpiece.

[0005] On the other hand, in the conventional double-sided polishing apparatuses described in Patent Document 1 and Patent Document 2, only the polishing conditions during the main polishing are set. That is, in the conventional double-sided polishing apparatus, during the main polishing, the distance between the surface plates is controlled, or the main polishing of the workpiece is stopped at the timing when the shape index of the entire workpiece reaches the set value. However, it is not considered that the shape of the workpiece also changes when the load applied to the workpiece after the main polishing is gradually reduced. Therefore, the shape of the workpiece at the end of the final polishing may not match the target shape, and it may not be possible to obtain the desired workpiece shape deviating from the target shape.

[0006] The present invention has been made paying attention to the above problems, and an object thereof is to provide a double-sided polishing apparatus and a double-sided polishing method capable of suppressing the deviation of the shape of the workpiece at the end of the final polishing from the target shape and making it possible to obtain the desired workpiece shape.

Means for Solving the Problems

[0007] To achieve the above object, the double-sided polishing apparatus of the present invention sandwiches a disk-shaped workpiece between a lower platen and an upper platen disposed opposite to the lower platen, and relatively moves the lower platen, the upper platen, and the workpiece with a load applied to the workpiece to polish the front and back surfaces of the workpiece; a thickness measuring device that measures the thickness of the workpiece during polishing of the workpiece by the polishing machine; and a control unit that controls the polishing machine based on the measurement result of the thickness measuring device. The double-sided polishing apparatus is characterized in that a predetermined range of an outer peripheral region extending radially inward from the outer peripheral end and a planar region within a range from the outer peripheral region to the center of the workpiece are set on the workpiece. The control unit includes a first learning device that learns the correlation between the polishing conditions of main polishing for polishing the workpiece while maintaining the load at a target load, the outer peripheral flatness that is the flatness of the outer peripheral region, and the in-plane flatness that is the flatness of the planar region; a second learning device that learns the correlation between the polishing conditions of end-point polishing for polishing the workpiece while gradually reducing the load, the degree of change in the in-plane flatness at the stop time of the main polishing, the degree of change in the outer peripheral flatness at the stop time of the main polishing, and the amount of change in the outer peripheral flatness during the end-point polishing; a main polishing condition correction unit that obtains a correction target value for the in-plane flatness based on the polishing conditions of the main polishing set at the correction time and the estimated value of the in-plane flatness obtained by inputting the outer peripheral flatness at the correction time into the first learning device, and corrects the polishing conditions of the main polishing set at the correction time according to the polishing conditions of the main polishing obtained by inputting the correction target value for the in-plane flatness and the target value for the outer peripheral flatness into the first learning device; and a main polishing stop condition setting unit that inputs the polishing conditions of the end-point polishing, the degree of change in the in-plane flatness at the calculation time, and the degree of change in the outer peripheral flatness at the calculation time into the second learning device to obtain the amount of change in the outer peripheral flatness during the end-point polishing, and sets the stop condition of the main polishing based on the amount of change in the outer peripheral flatness and the target value for the outer peripheral flatness.

[0008] In order to achieve the above object, the double-sided polishing method of the present invention sandwiches a disk-shaped workpiece between a lower platen and an upper platen disposed opposite to the lower platen, and while applying a load to the workpiece, the lower platen, the upper platen, and the workpiece are relatively moved to polish the front and back surfaces of the workpiece. In the double-sided polishing method, while maintaining the load at a target load, the polishing conditions for main polishing for polishing the workpiece, the outer peripheral flatness which is the flatness of an outer peripheral region set in a predetermined range from the outer peripheral end of the workpiece toward the radially inner side, and the in-plane flatness which is the flatness of an in-plane region in the range from the outer peripheral region of the workpiece to the workpiece center are learned by a first learning device. A second learning device learns the correlation between the polishing conditions for end polishing for polishing the workpiece while gradually reducing the load, the degree of change in the in-plane flatness at the stop time of the main polishing, the degree of change in the outer peripheral flatness at the stop time of the main polishing, and the amount of change in the outer peripheral flatness during the end polishing. Using these, a step of setting a target value for the outer peripheral flatness and a target value for the in-plane flatness, a step of performing initial polishing for polishing the workpiece while gradually increasing the load after setting the target values for the outer peripheral flatness and the in-plane flatness, a step of starting the main polishing after the initial polishing is completed, during the execution of the main polishing, based on the polishing conditions of the main polishing set at the correction time and the estimated value of the in-plane flatness obtained by inputting the outer peripheral flatness at the correction time into the first learning device, obtaining a correction target value for the in-plane flatness, and correcting the polishing conditions of the main polishing set at the correction time according to the polishing conditions of the main polishing obtained by inputting the correction target value for the in-plane flatness and the target value for the outer peripheral flatness into the first learning device. During the execution of the main polishing, inputting the polishing conditions for the end polishing, the degree of change in the in-plane flatness at the calculation time, and the degree of change in the outer peripheral flatness at the calculation time into the second learning device to obtain the amount of change in the outer peripheral flatness during the end polishing, and setting the stop condition for the main polishing based on the amount of change in the outer peripheral flatness and the target value for the outer peripheral flatness. After the stop condition for the main polishing is satisfied, a step of executing the end polishing is provided, which is characterized by this.

Effect of the Invention

[0009] In the double-sided polishing apparatus and the double-sided polishing method of the present invention, it is possible to suppress the deviation of the shape of the workpiece from the target shape at the end of the final polishing, and it is possible to obtain a desired workpiece shape.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3A

Figure 3B

Figure 4A

Figure 4B

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Embodiments for Carrying Out the Invention

[0011] Hereinafter, embodiments for carrying out the double-sided polishing apparatus and the double-sided polishing method of the present invention will be described based on Example 1 shown in the drawings.

[0012] The double-sided polishing apparatus 1 of Example 1 is a double-sided polishing apparatus for polishing both the front and back surfaces of a thin plate disk-shaped workpiece W such as a semiconductor wafer, a quartz wafer, a sapphire wafer, a glass wafer, or a ceramic wafer. As shown in FIG. 1, the double-sided polishing apparatus 1 includes a polishing machine 10, a thickness measuring device 20, and a control unit 30.

[0013] The polishing machine 10 sandwiches the workpiece W between a lower platen 11 and an upper platen 12 disposed opposite to the lower platen 11, and relatively moves the lower platen 11, the upper platen 12, and the workpiece W in a state where a load is applied to the workpiece W to simultaneously polish the front and back surfaces of the workpiece W. The polishing machine 10 includes a donut disk-shaped lower platen 11 and an upper platen 12 concentrically arranged with the axis L1 at the center, a sun gear 13 rotatably disposed at the center of the lower platen 11, an internal gear 14 disposed on the outer peripheral side of the lower platen 11, and a carrier plate 15 disposed between the lower platen 11 and the upper platen 12 and having a workpiece holding hole 15a (see FIG. 2). Further, a polishing pad 11a is attached to the upper surface of the lower platen 11, and a polishing pad 12a is attached to the lower surface of the upper platen 12. Furthermore, the upper platen 12 is provided with a plurality of supply holes (not shown) for supplying a polishing slurry (hereinafter referred to as "slurry"). In some cases, it may be possible to select the presence or absence of slurry supply for each supply hole or for each group of supply holes.

[0014] Here, as shown in FIG. 2, the carrier plate 15 meshes with the sun gear 13 and the internal gear 14. Then, as the sun gear 13 and the internal gear 14 rotate, the carrier plate 15 rotates (revolves) around the axis L1 while rotating on its own axis.

[0015] The workpiece W is disposed within the workpiece holding hole 15a of the carrier plate 15. Then, with the workpiece W sandwiched between the polishing pad 11a attached to the rotating lower platen 11 and the polishing pad 12a attached to the rotating upper platen 12, as the carrier plate 15 rotates on its own axis and revolves, the lower platen 11 and the upper platen 12 and the workpiece W move relative to each other, and the workpiece W is polished by the polishing pads 11a and 12a. That is, the surfaces of the polishing pads 11a and 12a become polishing surfaces for polishing the workpiece W.

[0016] The upper platen 12 is fixed to the rod 16 via the support studs 16a and the mounting members 16b attached to the upper surface. The rod 16 is expanded and contracted in the vertical direction by the fifth drive device M5, and the upper platen 12 moves up and down as the rod 16 expands and contracts. Then, a predetermined load is applied to the workpiece W from the upper platen 12 according to the expansion and contraction length of the rod 16. That is, the load applied to the workpiece W is adjusted by controlling the fifth drive device M5.

[0017] Also, at the center of the grinding machine 10, a first drive shaft 17a standing along the axis L1 is arranged. The first drive shaft 17a is a shaft that is rotated by a first drive device M1. A driver 18 is fixed to the upper end portion of the first drive shaft 17a. Thus, the driver 18 rotates integrally with the first drive shaft 17a. On the other hand, a groove portion (not shown) that engages with a hook 12b provided on the upper surface plate 12 is formed on the outer peripheral surface of the driver 18. Then, when the rod 16 extends and the upper surface plate 12 moves downward, and the hook 12b engages with the groove portion of the driver 18, the driver 18 and the upper surface plate 12 rotate integrally. That is, since the upper surface plate 12 rotates integrally with the first drive shaft 17a that is rotated by the first drive device M1, the rotation speed of the upper surface plate 12 is adjusted by controlling the first drive device M1.

[0018] A second drive shaft 17b is fixed in a penetrating state to a hole 13a at the center of the sun gear 13. The second drive shaft 17b is a hollow tube with both ends open, and the first drive shaft 17a penetrates it rotatably. Also, the second drive shaft 17b is rotated by a second drive device M2. That is, since the sun gear 13 rotates integrally with the second drive shaft 17b that is rotated by the second drive device M2, the rotation speed of the sun gear 13 is adjusted by controlling the second drive device M2.

[0019] A third drive shaft 17c is formed at the lower part of the center of the lower surface plate 11. The third drive shaft 17c is a hollow tube with both ends open, and the first drive shaft 17a and the second drive shaft 17b penetrate it rotatably. Also, the third drive shaft 17c is rotated by a third drive device M3. That is, since the lower surface plate 11 rotates integrally with the third drive shaft 17c that is rotated by the third drive device M3, the rotation speed of the lower surface plate 11 is adjusted by controlling the third drive device M3.

[0020] In addition, a fourth drive shaft 17d is formed on the internal gear 14. The fourth drive shaft 17d is a hollow tube with both ends open, and the first drive shaft 17a, the second drive shaft 17b, and the third drive shaft 17c rotatably penetrate through it. Also, the fourth drive shaft 17d is rotated by a fourth drive device M4. That is, since the internal gear 14 rotates integrally with the fourth drive shaft 17d rotated by the fourth drive device M4, the rotation speed of the internal gear 14 is adjusted by controlling the fourth drive device M4.

[0021] Then, the double-sided polishing apparatus 1 of the first embodiment polishes the workpiece W while controlling the load applied to the workpiece W by expanding and contracting the rod 16 based on predetermined polishing conditions, and controlling the rotation speeds of the lower fixed plate 11, the upper fixed plate 12, the sun gear 13, and the internal gear 14. Here, since it takes time to adjust (increase or decrease) the expansion and contraction of the rod 16 and the rotation speeds of the lower fixed plate 11 and the like, the polishing process of the workpiece W is divided into initial polishing, main polishing, and end-point polishing.

[0022] Note that "initial polishing" is a process of polishing the workpiece W while gradually increasing the load applied to the workpiece W after the start of polishing. In the initial polishing, the rotation speeds of the lower fixed plate 11, the upper fixed plate 12, the sun gear 13, and the internal gear 14 are also gradually increased respectively. And the polishing conditions (initial polishing conditions) of the initial polishing including the polishing time and end conditions in the initial polishing are defined by the preliminary polishing conditions. Note that the "preliminary polishing conditions" are polishing conditions determined in advance according to the final processing target (polishing target) of the workpiece W and the type of the workpiece W and the like.

[0023] Also, "main polishing" is a process of polishing the workpiece W under predetermined polishing conditions while maintaining the load applied to the workpiece W at the target load after the initial polishing is completed. In main polishing, the rotational speeds of the lower platen 11, upper platen 12, sun gear 13, and internal gear 14 are also maintained at their respective target rotational speeds. The polishing conditions of the main polishing (main polishing conditions) including the target load and the target rotational speeds of various rotational speeds in the main polishing are defined by the preliminary polishing conditions at the start of the main polishing and are continuously corrected repeatedly during the execution of the main polishing. Further, the stop conditions of the main polishing (main polishing stop conditions) that define the timing of stopping the main polishing are repeatedly defined during the execution of the main polishing, and the main polishing is stopped when the main polishing stop conditions are satisfied. Note that the main polishing may be composed of a plurality of sections (a plurality of steps), and the main polishing conditions may include appropriately adjusting the temperature, flow rate, supply destination, etc. of the slurry according to the monitoring results of the change tendency of the shape of the workpiece W.

[0024] And "finish polishing" is a process of polishing the workpiece W while gradually reducing the load applied to the workpiece W after the main polishing is stopped. In finish polishing, the rotational speeds of the lower platen 11, upper platen 12, sun gear 13, and internal gear 14 are also gradually reduced. The polishing conditions of the finish polishing (finish polishing conditions) including the polishing time in the finish polishing are defined by the preliminary polishing conditions. The finish polishing conditions may also include appropriately adjusting the temperature, flow rate, supply destination, etc. of the slurry according to the monitoring results of the change tendency of the shape of the workpiece W.

[0025] A measurement hole 19 is formed in the upper platen 12 at a position separated from the center by a predetermined distance along the radial direction. The measurement hole 19 penetrates the upper platen 12 and the polishing pad 12a, and a window member 19a that transmits laser light, which is measurement light, is attached thereto.

[0026] The thickness measuring device 20 irradiates the workpiece W with measurement light, receives the measurement light reflected by the workpiece W, and measures the thickness of the workpiece W being polished (the distance between the front and back surfaces of the workpiece W). Further, the thickness measuring device 20 of Example 1 numerically values the shape of the workpiece W from the measured thickness data group of the workpiece W. The thickness measuring device 20 includes a measurement unit 21, a thickness measurement unit 22, and a shape calculation unit 23.

[0027] The measurement unit 21 is attached to the upper surface plate 12 and rotates integrally with the upper surface plate 12. Further, the measurement unit 21 includes a laser light source (not shown) that irradiates the workpiece W with laser light, which is measurement light, through a window member 19a mounted in the measurement hole 19 of the upper surface plate 12, and a light receiving unit (not shown) that receives the reflected light reflected by the front and back surfaces of the workpiece W as a light reception signal. The light reception signal received by the light receiving unit is transmitted to the thickness measurement unit 22 by a transmission unit 21a.

[0028] The thickness measurement unit 22 measures the thickness of the workpiece W by, for example, the optical reflection interference method. This thickness measurement unit 22 has a reception unit 22a that receives the light reception signal transmitted from the measurement unit 21, and acquires the thickness data of the workpiece W based on the light reception signal received by the reception unit 22a.

[0029] Here, due to the rotation of the upper surface plate 12, as shown in FIG. 3A, during the period when the measurement hole 19 passes over the surface of the workpiece W, the laser light from the measurement unit 21 is continuously irradiated onto the surface of the workpiece W. Therefore, the thickness measurement unit 22 continuously measures the thickness at each in-plane position of the workpiece W on the passing trajectories Na to Nc of the measurement hole 19. Then, while the measurement hole 19 passes through each of the passing trajectories Na to Nc (during the passing period of the measurement hole 19 from one end W1a to W3a of the workpiece W to the other end W1b to W3b), the thickness measurement unit 22 outputs a thickness data group composed of a large number of continuous thickness data for each passing. As a result, each time the measurement hole 19 passes over the surface of the workpiece W, the thickness measurement unit 22 outputs a thickness data group composed of a plurality of continuous data obtained by measuring the thickness at each in-plane position of the workpiece W (see FIG. 3B). Note that the thickness data group output from the thickness measurement unit 22 is input to the shape calculation unit 23.

[0030] Also, when inputting the thickness data group, the correlation between the measurement data (e.g., GBIR, ESFQD, etc.) measured by an external measuring instrument (external measuring device) capable of measuring the thickness (shape) of the work W and the thickness data group may be separately obtained, and the input may be made taking into account correction values and the like obtained from the correlation. Note that "GBIR (Global Backside Ideal Range)" is a value indicating the difference between the maximum value and the minimum value of the thickness distribution. Also, "ESFQD (Edge Site Front Least Square)" is a value that outputs the larger of the absolute values of the maximum and minimum distances from the reference plane at the outer peripheral portion of the wafer.

[0031] Furthermore, the thickness data group and the numerical values of the in-plane flatness and outer peripheral flatness described later do not necessarily have a high correlation with the measurement data of the external measuring instrument. For this reason, a value obtained from a difference database in which the differences between the thickness data group and the measurement data of the external measuring instrument are stored in a database, or a value obtained by inputting the obtained value into a conversion formula, or a value obtained by inputting the measurement data measured by an external measuring instrument into a conversion formula, etc., may be used to correct the thickness data group for the thickness data group, in-plane flatness, and outer peripheral flatness at any time or at a desired timing.

[0032] Note that the difference database may be learned and updated at any time. Thereby, the correlation between the thickness data group measured by the thickness measurement unit 22 and the measurement data of the external measuring instrument can be grasped. For example, the target value of the work shape input to the control unit 30 can be input according to the measurement standard of the external measuring instrument.

[0033] The shape calculation unit 23 converts the shape (hereinafter referred to as "work shape") of the work W into numerical information based on the thickness data group of the work W measured by the thickness measurement unit 22. In the first embodiment, the work shape is indicated by the in-plane flatness and the outer peripheral flatness. Note that the information on the in-plane flatness and the outer peripheral flatness output from the shape calculation unit 23 is input to the control unit 30.

[0034] Here, the "in-plane flatness" refers to the flatness of the in-plane region G set in the workpiece W, and is taken as the slope value of the approximate straight line (the approximate straight line indicated by A in FIGS. 4A and 4B) set by, for example, the least squares method with respect to the thickness data group in the in-plane region G. Note that the approximate straight line indicated by A' in FIGS. 4A and 4B is also an approximate straight line set with respect to the thickness data group in the in-plane region G. However, when the workpiece shape is converted into numerical information in the shape calculation unit 23, when viewed from the cross-section of the workpiece W broken along a straight line passing through the workpiece center Wo, the regions on the right side of the workpiece center Wo (the in-plane region G' and the outer peripheral region E') are ignored.

[0035] Further, the approximate straight line is represented by the following formula (1), and the slope value of the approximate straight line is "a" in the following formula (1). Approximate straight line: Y = aX + b ···(1) a: Slope value b: Intercept

[0036] When the in-plane flatness is a positive value, the workpiece shape becomes a convex shape with the center of the in-plane region G protruding. When the in-plane flatness is a negative value, the workpiece shape becomes a concave shape with the center of the in-plane region G sunken.

[0037] Also, the "outer peripheral flatness" refers to the flatness of the outer peripheral region E set in the workpiece W. The outer peripheral flatness is obtained by performing a continuity process on the slope values of the approximate straight lines set for the thickness data group of the outer peripheral region E with respect to the in-plane flatness as a reference. Here, the "continuity process on the slope values of the approximate straight lines set for the thickness data group of the outer peripheral region E with respect to the in-plane flatness as a reference" refers to a process of adding or subtracting the slope value of the approximate straight line set for the thickness data group in the outer peripheral region E with respect to the in-plane flatness. That is, the "continuity process" includes a process of adding the in-plane flatness and the slope value of the approximate straight line set for the thickness data group in the outer peripheral region E (a process of calculating the sum value) and a process of calculating the difference between the in-plane flatness and the slope value of the approximate straight line set for the thickness data group in the outer peripheral region E (a process of calculating the difference value).

[0038] The outer peripheral flatness of Example 1 is obtained by adding the slope value of the approximate straight line (the straight line indicated by B in FIGS. 4A and 4B) set for the thickness data group in the outer peripheral region E to the in-plane flatness, and is the value of the sum of the in-plane flatness and the slope value of the approximate straight line in the outer peripheral region E.

[0039] When the outer peripheral flatness is a positive value, the work shape is a roll-up shape in which the outer peripheral region E jumps up with respect to the in-plane region G. When the outer peripheral flatness is a negative value, the work shape is a roll-off shape in which the outer peripheral region E sags with respect to the in-plane region G.

[0040] The "outer peripheral region E" is a region within a predetermined range radially inward from the outer peripheral edge We of the work W, as shown in FIGS. 3A and 3B. The "in-plane region G" is a region within the range from the radially inner edge of the outer peripheral region E to the work center Wo, as shown in FIGS. 3A and 3B. That is, as shown in FIG. 3A, the outer peripheral region E is an annular region between the outer peripheral edge We of the work W and the circle indicated by the broken line set within the work W, and is set at the peripheral edge of the work W. The in-plane region G is a circular region surrounded by the circle indicated by the broken line set within the work W, and is set at the central part of the work W. Then, as shown in FIG. 3B, when viewed in the cross-section of the work W cut by a straight line passing through the work center Wo, the in-plane region G and the outer peripheral region E are set symmetrically with the work center Wo in between. Also, the setting range of the outer peripheral region E can be arbitrarily determined.

[0041] FIG. 4A shows an approximate straight line created based on the thickness data group of the work W in the first example. In the case shown in FIG. 4A, the slope value of the approximate straight line A in the in-plane region G is +50 (nm: nanometer), and the slope value of the approximate straight line B in the outer peripheral region E is zero (nm). Therefore, the in-plane flatness of the work W in the first example is +50 (nm), and the outer peripheral flatness is +50 (= +50 + zero) (nm).

[0042] Further, FIG. 4B shows an approximate straight line created based on the thickness data group of the work W of the second example. In the case shown in FIG. 4B, the slope value of the approximate straight line A in the in-plane region G is +30 (nm), and the slope value of the approximate straight line B in the outer peripheral region E is -30 (nm). Therefore, the in-plane flatness of the work W of the second example is +30 (nm), and the outer peripheral flatness becomes zero (= +30 + (-30)) (nm).

[0043] Further, as shown in FIG. 5, the outer peripheral region E may be divided into a plurality (three in the example shown in FIG. 5) of sections along the radial direction of the work W. In this case, the outer peripheral flatness is calculated for each section Ei, Eo1, Eo2 of the outer peripheral region E.

[0044] That is, in the example shown in FIG. 5, among the outer peripheral region E, a predetermined section adjacent to the in-plane region G is defined as the innermost section Ei. Also, a section adjacent to the innermost section Ei and radially outside the innermost section Ei is defined as the first outer section Eo1. Further, a section adjacent to the first outer section Eo1 and radially outside the first outer section Eo1 is defined as the second outer section Eo2.

[0045] Then, the flatness of the innermost section Ei is obtained by performing a continuity process on the slope values of the approximate straight line set for the thickness data group of the innermost section Ei with respect to the in-plane flatness as a reference. Here, the flatness of the innermost section Ei is obtained by adding the slope value of the approximate straight line set for the thickness data group of the innermost section Ei to the in-plane flatness. That is, the flatness of the innermost section Ei in Example 1 is the sum of the in-plane flatness and the slope value of the approximate straight line of the innermost section Ei.

[0046] Further, the flatness of the first outer section Eo1 is obtained by performing a continuity process on the slope values of the approximate straight lines set for the thickness data group of the first outer section Eo1 with reference to the flatness of the section (the innermost section Ei) adjacent to the first outer section Eo1 on the radially inner side. Here, the flatness of the first outer section Eo1 is obtained by adding the slope value of the approximate straight line set for the thickness data group of the first outer section Eo1 to the flatness of the innermost section Ei. That is, the flatness of the first outer section Eo1 in the first embodiment is the sum of the flatness of the innermost section Ei and the slope value of the approximate straight line of the first outer section Eo1.

[0047] Also, the flatness of the second outer section Eo2 is obtained by performing a continuity process on the slope values of the approximate straight lines set for the thickness data group of the second outer section Eo2 with reference to the flatness of the section (the first outer section Eo1) adjacent to the second outer section Eo2 on the radially inner side. Here, the flatness of the second outer section Eo2 is obtained by adding the slope value of the approximate straight line set for the thickness data group of the second outer section Eo2 to the flatness of the first outer section Eo1. That is, the flatness of the second outer section Eo2 in the first embodiment is the sum of the flatness of the first outer section Eo1 and the slope value of the approximate straight line of the second outer section Eo2.

[0048] Furthermore, the shape calculation unit 23 continuously calculates numerical information (in-plane flatness and outer peripheral flatness) indicating the work shape at arbitrary intervals (for example, every 10 to 15 seconds) during the polishing of the work W. Thereby, the control unit 30 can respectively obtain the degree of change in the in-plane flatness and the degree of change in the outer peripheral flatness during polishing.

[0049] Here, the degree of change in the in-plane flatness is, for example, the slope value of the approximate straight line set by the least squares method for the in-plane flatness data group indicated by ○ in FIG. 6, that is, the ratio of the change in the in-plane flatness within a predetermined time width. Also, the degree of change in the outer peripheral flatness is, for example, the slope value of the approximate straight line set by the least squares method for the outer peripheral flatness data group indicated by ● in FIG. 6, that is, the ratio of the outer peripheral flatness within a predetermined time width. Then, the control unit 30 determines the change tendency of the work shape from the degree of change in the in-plane flatness and the degree of change in the outer peripheral flatness.

[0050] That is, when the degree of change in the in-plane flatness is a positive value, the control unit 30 determines that the change tendency of the in-plane region G tends to change in the convex direction. Also, when the degree of change in the in-plane flatness is a negative value, the control unit 30 determines that the change tendency of the in-plane region G tends to change in the concave direction. For example, in the example shown in FIG. 6, the degree of change in the in-plane flatness at time t0 (the slope value of the approximate straight line set for the data group of the in-plane flatness within a minute time width including time t0 as a reference) becomes a negative value as indicated by the dashed-dotted line. Therefore, it is determined that the change tendency of the in-plane region G after time t0 is a change tendency to change in the concave direction.

[0051] Also, when the degree of change in the outer peripheral flatness is a positive value, the control unit 30 determines that the change tendency of the outer peripheral region E tends to change in the roll-up (bouncing up) direction. Also, when the degree of change in the outer peripheral flatness is a negative value, the control unit 30 determines that the change tendency of the outer peripheral region E tends to change in the roll-off (sagging) direction. For example, in the example shown in FIG. 6, the degree of change in the outer peripheral flatness at time t0 (the slope value of the approximate straight line set for the data group of the outer peripheral flatness within a minute time width including time t0 as a reference) becomes a positive value as indicated by the two-dot chain line. Therefore, it is determined that the change tendency of the outer peripheral region E after time t0 is a change tendency to change in the roll-up direction.

[0052] The control unit 30 is composed of a CPU (Central Processing Unit) or the like, and includes a first learning device 31, a second learning device 32, a main polishing condition correction unit 33, a main polishing stop condition setting unit 34, a control arithmetic unit 35, a memory 36, and the like. Also, an input device 41 that can be operated by the operator of the double-sided polishing apparatus 1 and a display 42 that can be visually observed by the operator are connected to the control unit 30.

[0053] The control unit 30 outputs control commands from the control arithmetic unit 35 to the first to fifth drive devices M1 to M5 etc. based on the measurement result of the work W by the thickness measuring device 20 (which may include a correction value obtained by separately obtaining the correlation with the measurement data of the shape of the work W measured by an external measuring device), the processing target of the work W, the preliminary polishing conditions, conditions such as auxiliary materials, information regarding the device state of the polishing machine 10, the program stored in the memory 36, the main polishing conditions reset by the main polishing condition correction unit 33, the main polishing stop conditions set by the main polishing stop condition setting unit 34, etc., and controls the operation of the polishing machine 10. Note that the processing target of the work W, the preliminary polishing conditions, conditions such as auxiliary materials, and information regarding the device state of the polishing machine 10 may be input by the operator via the input device 41, or may be stored in the memory 36 in advance. Further, while controlling the operation of the polishing machine 10, the control unit 30 appropriately displays necessary information on the display 42.

[0054] The first learning device 31 is a learning device that learns the correlation between the main polishing conditions at an arbitrary timing during main polishing, the outer peripheral flatness acquired at that timing, and the in-plane flatness acquired at that timing. That is, the control unit 30 causes the first learning device 31 of the first embodiment to learn, for example, as shown in FIG. 7, the main polishing conditions at an arbitrary timing during main polishing, the outer peripheral flatness included in the main polishing conditions, disturbance factors, and the in-plane flatness acquired at the same timing as a data set. Here, the main polishing conditions include, for example, the rotation speed of the lower platen 11, the rotation speed of the upper platen 12, the rotation speed of the sun gear 13, the rotation speed of the internal gear 14, the revolution speed of the carrier plate 15, the rotation speed of the carrier plate 15, the load applied to the work W, the flow rate of the slurry, the type of the slurry, etc. Further, the disturbance factors include, for example, as the state of the auxiliary material, the usage period of the carrier plate 15 (carrier life), the usage period of the polishing pads 11a, 12a (pad life), dressing conditions, etc., and as the device state, the load factor of each drive device, the temperature of the area (drive chamber) where each drive device is arranged, the temperature of the polishing pads 11a, 12a, the temperature of the slurry, the fluctuation value of the load applied to the work W, etc. The main polishing conditions and the disturbance factors may be input by the operator via the input device 41, or may be detected by a sensor or the like.

[0055] The second learning device 32 is a learning device that learns the correlation between the end-point polishing conditions, the degree of change in the in-plane flatness at the time of main polishing stop, the degree of change in the outer peripheral flatness at the time of main polishing stop, and the amount of change in the outer peripheral flatness during end-point polishing. That is, as shown in FIG. 8, for example, the control unit 30 inputs a predetermined end-point polishing condition, the degree of change in the in-plane flatness at the time of main polishing stop, the degree of change in the in-plane flatness at the time of main polishing stop, and the amount of change in the outer peripheral flatness during end-point polishing performed under the end-point polishing condition on the premise of the degree of change to the second learning device 32 of the first embodiment as a data set for learning. Here, the end-point polishing conditions include, for example, the rotational speed of the lower platen 11, the rotational speed of the upper platen 12, the rotational speed of the sun gear 13, the rotational speed of the internal gear 14, the revolution speed of the carrier plate 15, the rotation speed of the carrier plate 15, the load applied to the workpiece W, the flow rate of the slurry, the type of the slurry, the end-point polishing time (deceleration time), and the like. The end-point polishing conditions may be input by the operator via the input device 41 or detected by a sensor or the like.

[0056] In addition, the "degree of change in the in-plane flatness at the time of main polishing stop" is the slope value of the approximate straight line set for the data group of the in-plane flatness within a minute time width including the main polishing stop time point (the timing of stopping the main polishing) as a reference. The "degree of change in the outer peripheral flatness at the time of main polishing stop" is the slope value of the approximate straight line set for the data group of the outer peripheral flatness within a minute time width including the main polishing stop time point as a reference. And the "amount of change in the outer peripheral flatness during end-point polishing" is the difference between the outer peripheral flatness at the main polishing stop time point and the outer peripheral flatness at the end-point polishing end time point.

[0057] The main polishing condition correction unit 33, during main polishing, based on the estimated value of the in-plane flatness obtained by inputting the main polishing conditions set at the correction time and the calculated value of the outer peripheral flatness at the correction time into the first learning device 31, obtains a correction target value for the in-plane flatness to make the calculated value of the in-plane flatness approach the target value. Then, the main polishing condition correction unit 33 corrects the main polishing conditions set at the correction time according to the correction target value of the in-plane flatness and the main polishing conditions obtained by inputting the final target value of the outer peripheral flatness at the end of polishing into the first learning device 31. That is, the main polishing condition correction unit 33 uses the first learning device 31 to obtain the main polishing conditions that satisfy the target value of the outer peripheral flatness and the correction target value, and replaces the main polishing conditions set at the correction time with the obtained main polishing conditions and re-sets them as new main polishing conditions. In the newly re-set main polishing conditions, for example, at least one of the rotation speed of the lower platen 11, the rotation speed of the upper platen 12, the rotation speed of the sun gear 13, the rotation speed of the internal gear 14, the revolution speed of the carrier plate 15, the rotation speed of the carrier plate 15, the load applied to the workpiece W, the flow rate of the slurry, the variety, the supply destination, etc. is changed.

[0058] Here, the "correction target value for the in-plane flatness to make the calculated value of the in-plane flatness approach the target value" is obtained by the following procedure. That is, the main polishing condition correction unit 33 first inputs the main polishing conditions at the correction time and the calculated value of the outer peripheral flatness at the correction time into the first learning device 31 to obtain an estimated value of the in-plane flatness. Next, the main polishing condition correction unit 33 calculates the difference value between the calculated value of the in-plane flatness at the correction time and the estimated value of the in-plane flatness. Then, the main polishing condition correction unit 33 sets the value calculated by subtracting the difference value of the in-plane flatness from the final target value of the in-plane flatness at the end of polishing as the "correction target value".

[0059] During the main polishing, the main polishing stop condition setting unit 34 inputs the end-point polishing condition, the degree of change in the in-plane flatness at the calculation time point, and the degree of change in the in-plane flatness at the calculation time point into the second learning device 32, and obtains the amount of change in the outer peripheral flatness during the end-point polishing. Then, based on the obtained amount of change in the outer peripheral flatness and the target value of the final outer peripheral flatness at the end of polishing, the stop condition of the main polishing (main polishing stop condition) is set. Here, the main polishing stop condition is defined by the outer peripheral flatness at the stop time of the main polishing.

[0060] That is, the main polishing stop condition setting unit 34 inputs the end-point polishing condition, the degree of change in the in-plane flatness at the calculation time point (the slope value of the approximate straight line set for the data group of the in-plane flatness within a minute time width including the calculation time point as a reference), and the degree of change in the outer peripheral flatness at the calculation time point (the slope value of the approximate straight line set for the data group of the outer peripheral flatness within a minute time width including the calculation time point as a reference) into the second learning device 32, and obtains the amount of change in the outer peripheral flatness during the end-point polishing in consideration of the degree of change. Next, the main polishing stop condition setting unit 34 calculates the target value of the outer peripheral flatness at the main polishing stop time by inverse calculation from the obtained amount of change in the outer peripheral flatness, the target value of the final outer peripheral flatness at the end of polishing, and the end-point polishing time. Then, the calculated target value of the outer peripheral flatness is defined as the main polishing stop condition.

[0061] During the execution of the initial polishing, the control unit 30 outputs a control command corresponding to the initial polishing condition set in advance under the premise polishing condition to the first to fifth driving devices M1 to M5 from the control arithmetic unit 35.

[0062] Also, at the start of the main polishing, the control unit 30 outputs a control command corresponding to the main polishing condition set in advance under the premise polishing condition to the first to fifth driving devices M1 to M5 from the control arithmetic unit 35.

[0063] During the execution of the main polishing, the control unit 30 corrects the main polishing condition by the first learning device 31 and the main polishing condition correction unit 33. When the main polishing condition is corrected, the control unit 30 outputs a control command corresponding to the corrected main polishing condition (the newly reset main polishing condition) to the first to fifth driving devices M1 to M5 from the control arithmetic unit 35.

[0064] Furthermore, during the execution of the main polishing, the control unit 30 sets the main polishing stop condition by the second learning device 32 and the main polishing stop condition setting unit 34. When the main polishing stop condition is satisfied, the control unit 30 outputs a control command to stop the main polishing and execute the end-point polishing to the first to fifth driving devices M1 to M5 from the control arithmetic unit 35.

[0065] Also, during the execution of the end-point polishing, the control unit 30 outputs a control command corresponding to the end-point polishing condition set in advance under the premise polishing condition to the first to fifth driving devices M1 to M5 from the control arithmetic unit 35.

[0066] FIG. 9 is a flowchart showing the flow of the polishing process executed by the control unit 30 of the double-sided polishing apparatus 1 of the first embodiment. Hereinafter, based on FIG. 9, the flow of the polishing process of the first embodiment will be described. The polishing process is executed in a state where the carrier plate 15 and the workpiece W are set on the polishing machine 10. Also, during the polishing process (from the start of the execution of the initial polishing to the end of the end-point polishing), the thickness measuring device 20 continuously inputs the shape information (information on the in-plane flatness and the outer peripheral flatness) of the workpiece W to the control unit 30.

[0067] In step S1, the control unit 30 sets the target value of the final in-plane flatness at the end of the end-point polishing, which is the final processing target of the workpiece W, and the target value of the final outer peripheral flatness at the end of the end-point polishing, and proceeds to step S2. Here, the target value of the in-plane flatness and the target value of the outer peripheral flatness are input by the operator via the input device 41. Note that the processing target of the workpiece W may be set in advance for each type of the workpiece W and stored in the memory 36. In this case, the control unit 30 reads and sets the processing target of the workpiece W from the memory 36 based on the type of the workpiece W input by the operator.

[0068] In step S2, following the setting of the workpiece processing target in step S1, the control unit 30 sets the preliminary polishing conditions and proceeds to step S3. Here, the preliminary polishing conditions are various conditions that are the basis for polishing the workpiece W, which are preset for each workpiece processing target and type of workpiece W. The preliminary polishing conditions include, for example, initial polishing conditions including the polishing time and polishing end conditions in the initial polishing, main polishing conditions at the start of the main polishing, end-point polishing conditions including the polishing time and polishing end conditions in the end-point polishing, information regarding the slurry such as the slurry flow rate, slurry type, slurry supply destination, and information regarding the state of consumable auxiliary materials such as the carrier life. The preliminary polishing conditions are input by the operator via the input device 41 or read from the memory 36.

[0069] In step S3, following the setting of the preliminary polishing conditions in step S2 or the determination of non-completion of the initial polishing in step S4, the control unit 30 outputs a control command corresponding to the initial polishing conditions defined by the preliminary polishing conditions from the control arithmetic unit 35 to the first to fifth drive devices M1 to M5, executes the initial polishing, and proceeds to step S4. During the execution of the initial polishing, in the polishing machine 10, while gradually increasing the load applied to the workpiece W, the workpiece W is polished while gradually increasing the rotational speeds of the lower platen 11 and the upper platen 12, the rotational speed of the sun gear 13, and the rotational speed of the internal gear 14.

[0070] In step S4, following the execution of the initial polishing in step S3, the control unit 30 determines whether the initial polishing has ended. Then, if the control unit 30 determines YES (initial polishing ended), it proceeds to step S5, and if it determines NO (initial polishing not ended), it returns to step S3. Here, the control unit 30 determines that the initial polishing has ended when the end conditions of the initial polishing defined by the preliminary polishing conditions are satisfied, such as when the load applied to the workpiece W reaches a predetermined load or when the rotational speed of the lower platen 11 or the like reaches a predetermined rotational speed.

[0071] In step S5, following the determination of the end of the initial polishing in step S4, the control unit 30 outputs control commands corresponding to the main polishing conditions defined by the prerequisite polishing conditions from the control arithmetic unit 35 to the first to fifth drive devices M1 to M5, starts the execution of the main polishing, and proceeds to step S6 and step S7. During the execution of the main polishing, in the polishing machine 10, the load applied to the workpiece W is adjusted to the target load defined by the main polishing conditions, and the workpiece W is polished while adjusting the rotation speeds of the lower surface plate 11 and the upper surface plate 12, the rotation speed of the sun gear 13, and the rotation speed of the internal gear 14 to their respective target rotation speeds.

[0072] In step S6, following the start of the execution of the main polishing in step S5, the control unit 30 corrects the main polishing conditions by the main polishing condition correction unit 33. The correction of the main polishing conditions is executed in parallel with the setting of the main polishing stop conditions in step S7 described later, and is repeatedly executed at regular intervals (for example, intervals of about 300 seconds) until it is determined in step S8 described later that the main polishing has stopped.

[0073] Here, the main polishing condition correction unit 33 corrects the main polishing conditions according to the following procedure. (1) Input the main polishing conditions at the correction time and the calculated value of the outer peripheral flatness at the correction time into the first learning device 31, and obtain the estimated value of the in-plane flatness at the correction time under the main polishing conditions. (2) Subtract the estimated value of the in-plane flatness obtained in (1) from the calculated value of the in-plane flatness at the correction time to calculate the difference value of the in-plane flatness (the difference between the estimated value and the current value of the in-plane flatness). (3) Calculate the value obtained by subtracting the difference value of the in-plane flatness calculated in (2) from the target value of the in-plane flatness set in step S1, and set it as the "correction target value". (4) Input the "correction target value" set in (3) and the target value of the outer peripheral flatness set in step S1 into the first learning device 31, and obtain the main polishing conditions. (5) Replace the main polishing conditions obtained in (4) with the main polishing conditions set at the correction time and reset them as new main polishing conditions to correct the main polishing conditions.

[0074] In step S7, following the start of execution of the main polishing in step S5, the control unit 30 sets the main polishing stop condition by the main polishing stop condition setting unit 34. Note that the setting of the main polishing stop condition is executed in parallel with the correction of the main polishing condition in step S6 as described above, and is repeatedly executed at regular intervals (for example, intervals of about 1 second) until it is determined in step S8 described later that the main polishing should be stopped.

[0075] Here, the main polishing stop condition setting unit 34 sets the main polishing stop condition according to the following procedure. (1) Input the end point polishing condition determined by the preliminary polishing condition set in step S2, the degree of change in the in-plane flatness at the calculation time point, and the degree of change in the outer peripheral flatness at the calculation time point into the second learning device 32, and obtain the change amount of the outer peripheral flatness during the end point polishing (note that the "calculation time point" here is the time point when the main polishing stop condition is set). (3) Calculate the outer peripheral flatness at the main polishing stop time point (the target value of the outer peripheral flatness at the main polishing stop time point) required for the outer peripheral flatness to satisfy the final target value at the end of polishing by inverse calculation from the change amount of the outer peripheral flatness during the end point polishing obtained in step S1, the target value of the outer peripheral flatness set in step S1, and the polishing time of the end point polishing. (4) Set the outer peripheral flatness obtained in (3) as the main polishing stop condition. Note that the calculation of the outer peripheral flatness in (3) is continuously performed at regular intervals. Therefore, the main polishing stop condition is updated each time the outer peripheral flatness is calculated.

[0076] In step S8, following the setting of the main polishing stop condition in step S7, the control unit 30 determines whether to stop the main polishing. Then, when the control unit 30 determines YES (stop main polishing), it proceeds to step S9, and when it determines NO (continue main polishing), it returns to step S7. Here, the main polishing stop condition is that the calculated value of the outer peripheral flatness has reached the outer peripheral flatness set as the main polishing stop condition in step S7. Therefore, in step S8, it is determined whether the current outer peripheral flatness matches the outer peripheral flatness obtained in step S7.

[0077] In step S9, following the determination of the main polishing stop in step S8, the control unit 30 stops the main polishing, outputs a control command corresponding to the end-point polishing conditions defined by the premise polishing conditions to the first to fifth drive devices M1 to M5 from the control arithmetic unit 35, starts the execution of the end-point polishing, and proceeds to step S10. During the execution of the end-point polishing, in the polishing machine 10, the load applied to the workpiece W is gradually reduced, and the workpiece W is polished while reducing the rotation speeds of the lower platen 11 and the upper platen 12, the rotation speed of the sun gear 13, and the rotation speed of the internal gear 14, respectively.

[0078] In step S10, following the start of the execution of the end-point polishing in step S9, the control unit 30 determines whether the end-point polishing has ended. Then, if the control unit 30 determines YES (end-point polishing ended), it proceeds to step S11, and if it determines NO (end-point polishing not ended), it returns to step S9. Here, the control unit 30 determines that the end-point polishing has ended when the end-point polishing time has elapsed, when the load applied to the workpiece W has become equal to or less than a predetermined value, or when the end-condition of the end-point polishing defined by the premise polishing conditions is satisfied.

[0079] In step S11, following the determination of the end of the end-point polishing in step S10, the control unit 30 ends the polishing of the workpiece W by the polishing machine 10, records various polishing data in the memory 36, and proceeds to the end.

[0080] Hereinafter, the operation of the double-sided polishing apparatus 1 of Example 1 will be described by dividing it into a "correction control operation of the main polishing conditions", a "control operation of the main polishing stop conditions", and an "other control operation".

[0081] [Correction control operation of the main polishing conditions] In the double-sided polishing apparatus 1 of Example 1, during the main polishing, the in-plane flatness and the outer peripheral flatness change respectively. Therefore, if the main polishing conditions are not set in consideration of the correlation between the in-plane flatness and the outer peripheral flatness, it is difficult to accurately control both the in-plane flatness and the outer peripheral flatness.

[0082] That is, only the correlation between the in-plane flatness when the outer peripheral flatness is zero (nm) and the main polishing conditions in that case is learned by a learning device (hereinafter referred to as the "learning device of the comparative example"), and the target value of the in-plane flatness is input to the learning device of the comparative example to obtain the main polishing conditions. In a double-sided polishing apparatus (hereinafter referred to as the "double-sided polishing apparatus of the comparative example"), the state of the outer peripheral flatness is not considered. For this reason, an appropriate workpiece shape cannot be obtained.

[0083] Specifically, for example, assume that there is a workpiece W (hereinafter referred to as "exemplary workpiece W'") whose shape change of the workpiece for each main polishing condition is as shown in FIG. 10. That is, when the main polishing is performed under the main polishing condition A for the exemplary workpiece W', the in-plane flatness becomes +80 (nm) when the outer peripheral flatness is zero (nm), the in-plane flatness becomes +90 (nm) when the outer peripheral flatness is -20 (nm), and the in-plane flatness becomes +100 (nm) when the outer peripheral flatness is -50 (nm). Further, when the main polishing is performed under the main polishing condition B, the in-plane flatness becomes zero (nm) when the outer peripheral flatness is zero (nm), the in-plane flatness becomes +10 (nm) when the outer peripheral flatness is -20 (nm), and the in-plane flatness becomes +20 (nm) when the outer peripheral flatness is -50 (nm). Further, when the main polishing is performed under the main polishing condition C, the in-plane flatness becomes -20 (nm) when the outer peripheral flatness is zero (nm), the in-plane flatness becomes zero (nm) when the outer peripheral flatness is -20 (nm), and the in-plane flatness becomes +10 (nm) when the outer peripheral flatness is -50 (nm).

[0084] On the other hand, in the double-sided polishing apparatus of the comparative example as described above, only the correlation between the in-plane flatness when the outer peripheral flatness is zero (nm) and the main polishing conditions is learned by the learning device of the comparative example. For this reason, as shown in FIG. 11, the data set learned by the learning device of the comparative example is that the in-plane flatness is +80 (nm) for the main polishing condition A, the in-plane flatness is zero (nm) for the main polishing condition B, and the in-plane flatness is -20 (nm) for the main polishing condition C.

[0085] Here, consider the case where, during the main polishing, the exemplary workpiece W´ is polished so that the in-plane flatness becomes zero (nm) when the outer peripheral flatness is -20 (nm). At this time, in the double-sided polishing apparatus of the comparative example, since the outer peripheral flatness cannot be considered, after ignoring the outer peripheral flatness, when the target value of the in-plane flatness (here, zero (nm)) is input to the learning device of the comparative example, the result of "main polishing condition B" will be obtained. However, when the main polishing of the exemplary workpiece W´ is performed by setting the main polishing condition B, in the double-sided polishing apparatus of the comparative example, it is assumed that the in-plane flatness becomes +10 (nm) at the timing when the outer peripheral flatness is -20 (nm) during the main polishing. Also, if the in-plane flatness is made zero (nm), it is assumed that the outer peripheral flatness becomes zero (nm). That is, in the double-sided polishing apparatus of the comparative example, both the in-plane flatness and the outer peripheral flatness cannot be accurately made to match the target values.

[0086] On the other hand, the double-sided polishing apparatus 1 of Example 1 includes a first learning device 31 and a main polishing condition correction unit 33. Here, the first learning device 31 is a learning device that learns the correlation between the main polishing conditions, the outer peripheral flatness, and the in-plane flatness. That is, as shown in FIG. 10, the dataset to be learned by the first learning device 31 of Example 1 includes various main polishing conditions and the outer peripheral flatness and in-plane flatness under those conditions. Also, the main polishing condition correction unit 33 corrects the main polishing conditions during the main polishing based on the correction target value obtained based on the estimated value of the in-plane flatness obtained by inputting the main polishing conditions set at the correction time and the calculated value of the outer peripheral flatness at the correction time to the first learning device 31, and the main polishing conditions obtained by inputting the final target value of the outer peripheral flatness to the first learning device 31.

[0087] That is, in the double-sided polishing apparatus 1 of Example 1, during the execution of the main polishing, the main polishing conditions at the correction time and the calculated value of the outer peripheral flatness at the correction time are input to the first learning device 31. Next, the double-sided polishing apparatus 1 of Example 1 obtains an estimated value of the in-plane flatness at the correction time from the input information to the first learning device 31 and the learning result of the first learning device 31, and calculates a difference value from the actual in-plane flatness at the correction time. Then, the double-sided polishing apparatus 1 of Example 1 subtracts the difference value from the target value of the in-plane flatness to obtain a "correction target value". Then, the "correction target value" of the in-plane flatness and the target value of the outer peripheral flatness are input to the first learning device 31 to obtain new main polishing conditions, and the main polishing conditions set at the correction time are corrected (step S6).

[0088] Hereinafter, assuming that the dataset to be learned by the first learning device 31 is the content shown in FIG. 10, in the double-sided polishing apparatus 1 of Example 1, taking as an example the main polishing of a predetermined workpiece W with the goal of achieving an in-plane flatness of zero (nm) and an outer peripheral flatness of zero (nm) at the end of the final polishing, a specific description will be given.

[0089] In this case, since the target values at the end of the final polishing are an in-plane flatness of zero (nm) and an outer peripheral flatness of zero (nm), the main polishing is started by setting the main polishing conditions to B. Then, in the case of correcting the main polishing conditions at the timing when the calculated value of the outer peripheral flatness becomes -20 (nm) during the main polishing under the main polishing conditions B, the "main polishing conditions B", which are the main polishing conditions at the correction time, and the "-20 (nm)", which is the calculated value of the outer peripheral flatness at the correction time, are input to the first learning device 31 (see FIG. 10), and the estimated value of the in-plane flatness obtained as a result becomes "+10 (nm)".

[0090] On the other hand, assume that due to factors such as disturbances, the calculated value of the in-plane flatness (actual in-plane flatness) is +30 (nm).

[0091] In this case, since the estimated value of the in-plane flatness is "+10 (nm)" while the calculated value of the in-plane flatness is "+30 (nm)", it can be inferred that the actual in-plane flatness during the main polishing becomes a value +20 (nm) larger than the learning result of the first learning device 31.

[0092] Therefore, in the double-sided polishing apparatus 1 of Example 1, a difference value (+20 (nm)) between the estimated value of the in-plane flatness (“+10 (nm)”) and the calculated value of the in-plane flatness (“+30 (nm)”) is calculated. Then, the difference value (+20 (nm)) is subtracted from the target value of the in-plane flatness (zero (nm)) to obtain a “corrected target value (-20 (nm))”. Then, the double-sided polishing apparatus 1 of Example 1 inputs the corrected target value (-20 (nm)) of the in-plane flatness and the target value of the outer peripheral flatness (zero (nm)) to the first learning device 31 (see FIG. 10). As a result, a new main polishing condition “main polishing condition C” will be obtained. Then, instead of the main polishing condition (main polishing condition B) set at the correction time, “main polishing condition C” is set as the new main polishing condition.

[0093] If the main polishing condition is corrected to the main polishing condition C and the main polishing is executed, according to the learning result of the first learning device 31, when the outer peripheral flatness is zero (nm), the in-plane flatness is estimated to be -20 (nm). However, as described above, it is presumed that the actual in-plane flatness during the main polishing becomes a value +20 (nm) larger than the learning result of the first learning device 31. Therefore, the double-sided polishing apparatus 1 can make the in-plane flatness zero (nm) when the outer peripheral flatness is zero (nm). Thus, in the double-sided polishing apparatus 1 of Example 1, during the execution of the main polishing, the main polishing condition can be corrected in consideration of the correlation between the in-plane flatness and the outer peripheral flatness, and both the in-plane flatness and the outer peripheral flatness can be accurately controlled.

[0094] “Control Action of Main Polishing Stop Condition” In the double-sided polishing apparatus 1 of Example 1, it has been found that the workpiece shape, particularly the outer peripheral flatness, changes due to end-point polishing. Therefore, even if the workpiece shape reaches the target shape at the time of stopping the main polishing, the outer peripheral flatness may change due to end-point polishing, and the final workpiece shape may deviate from the target shape. Moreover, if the degree of change in the workpiece at the time of stopping the main polishing is different, even if end-point polishing is performed under the same end-point polishing conditions, the amount of change in the outer peripheral flatness during end-point polishing will be different. As a result, the final workpiece shape will be different from the workpiece shape at the time of stopping the main polishing, and it is assumed that it will deviate significantly from the target shape. Therefore, it is necessary to stop the main polishing in consideration of the amount of change in the outer peripheral flatness, taking into account the degree of change in the workpiece shape.

[0095] On the other hand, the double-sided polishing apparatus 1 of Example 1 includes a second learning device 32 and a main polishing stop condition setting unit 34. Here, the second learning device 32 is a learning device that learns the correlation between the end-point polishing conditions, the degree of change in the in-plane flatness at the time of stopping the main polishing, the degree of change in the outer peripheral flatness at the time of stopping the main polishing, and the amount of change in the outer peripheral flatness during end-point polishing. Further, the main polishing stop condition setting unit 34 sets, as the stop condition of the main polishing (main polishing stop condition), that the outer peripheral flatness of the workpiece W becomes the target outer peripheral flatness calculated based on the end-point polishing conditions, the degree of change in the in-plane flatness at the time of calculation, the degree of change in the outer peripheral flatness at the time of calculation, the amount of change in the outer peripheral flatness during end-point polishing obtained by inputting these to the second learning device 32, and the target value of the outer peripheral flatness at the end of end-point polishing.

[0096] That is, in the double-sided polishing apparatus 1 of Example 1, when setting the main polishing stop condition, the endpoint polishing condition, the degree of change in the in-plane flatness at the calculation time point, and the degree of change in the outer peripheral flatness at the calculation time point are input to the second learning device 32. Then, the double-sided polishing apparatus 1 of Example 1 obtains the change amount of the outer peripheral flatness during endpoint polishing from the input information to the second learning device 32 and the learning result of the second learning device 32. Then, from the change amount of the outer peripheral flatness during endpoint polishing, the target value of the outer peripheral flatness at the main polishing stop time point (target outer peripheral flatness) necessary to satisfy the target value of the outer peripheral flatness is obtained by inverse calculation from the target value of the outer peripheral flatness and the endpoint polishing time. And the main polishing stop condition is defined by the target outer peripheral flatness (step S7).

[0097] In this way, in the double-sided polishing apparatus 1 of Example 1, the outer peripheral flatness at the stop time of the main polishing can be stopped so that it becomes a necessary shape considering the change amount of the outer peripheral flatness by endpoint polishing based on the degrees of change in the in-plane flatness and the outer peripheral flatness. Therefore, the double-sided polishing apparatus 1 of Example 1 can prevent the work shape from deviating greatly from the target shape at the end point of the endpoint polishing (the final polishing end point) even if a shape change of the work W occurs due to the endpoint polishing.

[0098] Note that FIG. 12 shows the results of summarizing the shapes of the work W after performing endpoint polishing with the target value of the outer peripheral flatness at the end of polishing set to zero (nm) in the double-sided polishing apparatus 1 of Example 1. As shown in FIG. 12, in the first sample work, the in-plane flatness was +34 (nm) and the outer peripheral flatness was -2 (nm). In the second sample work, the in-plane flatness was -4 (nm) and the outer peripheral flatness was -5 (nm). In the third sample work, the in-plane flatness was +40 (nm) and the outer peripheral flatness was -3 (nm). In the fourth sample work, the in-plane flatness was +62 (nm) and the outer peripheral flatness was -3 (nm).

[0099] From these results, it can be seen that the double-sided polishing apparatus 1 of Example 1 can reduce the difference between the actual outer peripheral flatness and the target value (zero (nm)) regardless of the magnitude of the in-plane flatness. Therefore, it can be understood that the double-sided polishing apparatus 1 of Example 1 can prevent the work shape from deviating significantly from the target shape at the end point of the final polishing (the final polishing end point).

[0100] Also, FIG. 13 shows the polishing results of the work W at the final polishing end point when the work W is polished using the double-sided polishing apparatus 1 of Example 1. As shown in FIG. 13, in the fifth sample work, when the target value of the in-plane flatness was set to zero (nm) and the target value of the outer peripheral flatness was set to zero (nm), as a result of actual polishing, the in-plane flatness became -1 (nm) and the outer peripheral flatness became +3 (nm). Also, in the sixth sample work, when the target value of the in-plane flatness was set to zero (nm) and the target value of the outer peripheral flatness was set to -20 (nm), as a result of actual polishing, the in-plane flatness became -4 (nm) and the outer peripheral flatness became -20 (nm). Also, in the seventh sample work, when the target value of the in-plane flatness was set to -30 (nm) and the target value of the outer peripheral flatness was set to zero (nm), as a result of actual polishing, the in-plane flatness became -30 (nm) and the outer peripheral flatness became +2 (nm). Also, in the eighth sample work, when the target value of the in-plane flatness was set to +30 (nm) and the target value of the outer peripheral flatness was set to zero (nm), as a result of actual polishing, the in-plane flatness became +30 (nm) and the outer peripheral flatness became +3 (nm). Also, in the ninth sample work, when the target value of the in-plane flatness was set to zero (nm) and the target value of the outer peripheral flatness was set to zero (nm), as a result of actual polishing, the in-plane flatness became -5 (nm) and the outer peripheral flatness became -4 (nm).

[0101] From the results shown in FIG. 13, it became clear that the double-sided polishing apparatus 1 of Example 1 does not deviate significantly from the target values for both the in-plane flatness and the outer peripheral flatness, and can suppress the work shape at the end of the final polishing from deviating from the target shape.

[0102] [Other control actions] In the double-sided polishing apparatus 1 of Example 1, an in-plane region G and an outer peripheral region E are set for the workpiece W, and the workpiece shape is numerically represented by the in-plane flatness and the outer peripheral flatness. That is, in Example 1, the workpiece shape is divided into a plurality of regions and controlled. On the other hand, the workpiece W polished by the double-sided polishing apparatus 1 of Example 1 may be used as a substrate of a semiconductor element having fine electronic circuits formed on its surface. Here, the electronic circuits formed on the surface of the workpiece W may extend across the boundary between the in-plane region G and the outer peripheral region E. Therefore, it is desirable that the boundary is a smooth surface so that an inflection point is not formed at the boundary between the in-plane region G and the outer peripheral region E. In order to control the workpiece shape so that no inflection point is formed at the boundary between the in-plane region G and the outer peripheral region E, it is necessary to numerically represent the flatness (smoothness of the workpiece shape) on the boundary between different regions such as the in-plane region G and the outer peripheral region E, that is, the change in the slope of the workpiece shape at the boundary between the in-plane region G and the outer peripheral region E.

[0103] On the other hand, in the double-sided polishing apparatus 1 of Example 1, the in-plane flatness is set as the slope value of the approximate straight line set for the thickness data group of the in-plane region G. Further, the outer peripheral flatness is obtained by performing a continuity process on the slope values of the approximate straight lines set for the thickness data group of the outer peripheral region E with respect to the in-plane flatness as a reference. That is, in the double-sided polishing apparatus 1 of Example 1, the outer peripheral flatness is set as the sum value of the slope value of the approximate straight line set for the thickness data group of the outer peripheral region E and the in-plane flatness.

[0104] In this way, the double-sided polishing apparatus 1 of Example 1 calculates the in-plane flatness and the outer-periphery flatness separately. By obtaining the outer-periphery flatness through performing a continuity process on the slope values of the approximate straight lines set for the thickness data group of the outer-periphery region E with respect to the in-plane flatness as a reference, using the slope value of the approximate curve set for the data group of the in-plane region G as a reference line, the slope value of the approximate curve set for the data group of the outer-periphery region E with respect to the reference line can be quantified as the outer-periphery flatness. Therefore, the outer-periphery flatness can be shown as an index representing the degree of roll-off with respect to the in-plane flatness, and the flatness (smoothness of the workpiece shape) at the boundary between different regions such as the in-plane region G and the outer-periphery region E can be numerically expressed. Also, the double-sided polishing apparatus 1 of Example 1 can numerically indicate the shape of the workpiece W, making it easier for an operator or the like to grasp the workpiece shape.

[0105] Note that the outer-periphery flatness may be obtained by subtracting the slope value of the approximate straight line set for the thickness data group of the outer-periphery region E from the in-plane flatness. That is, the outer-periphery flatness may be the value of the difference between the in-plane flatness and the slope value of the approximate straight line set for the thickness data group of the outer-periphery region E.

[0106] Furthermore, the double-sided polishing apparatus 1 of Example 1 may divide the outer-periphery region E into a plurality of sections along the radial direction of the workpiece W, and calculate the outer-periphery flatness for each of the divided outer-periphery regions (the innermost section Ei, the outer section Eo). By dividing the outer-periphery region E into a plurality of sections, the double-sided polishing apparatus 1 of Example 1 can correct the main polishing conditions or set the main polishing stop conditions according to the shape change of each section. Thereby, the workpiece shape at the end of the final polishing can be made to match the target shape with higher accuracy.

[0107] Also, in the double-sided polishing apparatus 1 of Example 1, when dividing the outer-periphery region E into a plurality of sections, the section adjacent to the in-plane region G is set as the innermost section Ei, and the section radially outside the innermost section Ei is set as the outer section Eo.

[0108] Then, the flatness of the innermost section Ei is obtained by performing a continuity process on the slope values of the approximate straight lines set for the thickness data group of the innermost section Ei based on the in-plane flatness. The flatness of the innermost section Ei in Example 1 is the sum of the slope value of the approximate straight line set for the thickness data group of the innermost section Ei and the in-plane flatness. Also, the flatness of the outer section Eo is obtained by performing a continuity process on the slope values of the approximate straight lines set for the thickness data group of the outer section Eo with reference to the flatness of the section adjacent to the inner side in the radial direction of the outer section Eo (for example, the innermost section Ei). The flatness of the outer section Eo in Example 1 is the sum of the slope value of the approximate straight line set for the thickness data group of the outer section Eo and the flatness of the section adjacent to the inner side in the radial direction of the outer section Eo (for example, the innermost section Ei).

[0109] Accordingly, in the double-sided polishing apparatus 1 of Example 1, even when the outer peripheral region E is divided into a plurality of sections, the flatness (smoothness of the workpiece shape) of the outer peripheral region E can be appropriately expressed, and the workpiece shape can be accurately controlled.

[0110] Note that the flatness of the innermost section Ei may be obtained by subtracting the slope value of the approximate straight line set for the thickness data group of the innermost section Ei from the in-plane flatness. That is, the flatness of the innermost section Ei may be the difference value between the in-plane flatness and the slope value of the approximate straight line set for the thickness data group of the innermost section Ei.

[0111] Also, the flatness of the outer section Eo may be obtained by subtracting the slope value of the approximate straight line set for the thickness data group of the outer section Eo from the flatness of the section adjacent to the inner side in the radial direction of the outer section Eo (for example, the innermost section Ei). That is, the flatness of the outer section Eo may be the difference value between the flatness of the section adjacent to the inner side in the radial direction of the outer section Eo (for example, the innermost section Ei) and the slope value of the approximate straight line set for the thickness data group of the outer section Eo.

[0112] In addition, the outer section Eo of Example 1 is divided into two sections, a first outer section Eo1 and a second outer section Eo2. However, the outer section Eo may not be divided, or may be divided into three or more sections.

[0113] As described above, the double-sided polishing apparatus of the present invention has been described based on Example 1. However, the specific configuration is not limited to this example, and design changes and additions are allowed as long as they do not depart from the gist of the invention according to each claim.

[0114] In the double-sided polishing apparatus 1 of Example 1, in step S1 of the polishing process control shown in FIG. 9, when setting the target value of the in-plane flatness and the target value of the outer peripheral flatness, an example is shown in which the target value of the final in-plane flatness at the end of the final polishing, which is the final processing target of the workpiece W, and the target value of the final outer peripheral flatness at the end of the final polishing are set. However, the target value of the in-plane flatness and the target value of the outer peripheral flatness are not limited to this. For example, a plurality of target values corresponding to the polishing status of the workpiece W and the like may be set, and the target values may be appropriately varied when correcting the main polishing conditions and when setting the main polishing stop conditions.

Explanation of symbols

[0115] 1 Double-sided polishing apparatus 10 Polishing machine 11 Lower platen 12 Upper platen 20 Thickness measuring device 30 Control unit 31 First learning device 32 Second learning device 33 Main polishing condition correction unit 34 Main polishing stop condition setting unit G In-plane area E Outer peripheral area W Workpiece

Claims

1. A polishing machine that sandwiches a disc-shaped workpiece between a lower platen and an upper platen arranged opposite to the lower platen, and relatively moves the lower platen, the upper platen, and the workpiece with a load applied to the workpiece to polish the front and back surfaces of the workpiece; A thickness measuring device that measures the thickness of the workpiece during polishing by the polishing machine; A double-sided polishing apparatus comprising: a control unit that controls the polishing machine based on the measurement result of the thickness measuring device, wherein a predetermined range of an outer peripheral region extending from the outer peripheral end toward the radially inner side and a planar region within a range from the outer peripheral region to the center of the workpiece are set on the workpiece; the control unit a first learning device that learns the correlation between the polishing conditions of main polishing for polishing the workpiece while maintaining the load at a target load, the outer peripheral flatness that is the flatness of the outer peripheral region, and the in-plane flatness that is the flatness of the in-plane region; a second learning device that learns the correlation between the polishing conditions of end-point polishing for polishing the workpiece while gradually reducing the load, the degree of change in the in-plane flatness at the stop time of the main polishing, the degree of change in the outer peripheral flatness at the stop time of the main polishing, and the amount of change in the outer peripheral flatness during the end-point polishing; Based on the polishing conditions of the main polishing set at the correction time and the estimated value of the in-plane flatness obtained by inputting the outer peripheral flatness at the correction time into the first learning device, a correction target value of the in-plane flatness is obtained, and the correction target value of the in-plane flatness and the target value of the outer peripheral flatness are input into the first learning device to correct the polishing conditions of the main polishing set at the correction time by the obtained polishing conditions of the main polishing, a main polishing condition correction unit; The polishing conditions of the end-point polishing, the degree of change in the in-plane flatness at the calculation time, and the degree of change in the outer peripheral flatness at the calculation time are input into the second learning device to obtain the amount of change in the outer peripheral flatness during the end-point polishing, and based on the amount of change in the outer peripheral flatness and the target value of the outer peripheral flatness, a main polishing stop condition setting unit that sets the stop condition of the main polishing; A double-sided polishing apparatus, characterized by comprising the above.

2. In the double-sided polishing apparatus according to Claim 1, the in-plane flatness is the slope value of an approximate straight line set for a thickness data group in the in-plane region, and the outer peripheral flatness is obtained by performing a continuity process on the slope values of approximate straight lines set for a thickness data group in the outer peripheral region with respect to the in-plane flatness as a reference. A double-sided polishing apparatus, characterized by the above.

3. In the double-sided polishing apparatus according to claim 1 or claim 2, the outer peripheral region is divided into a plurality of sections along the radial direction of the workpiece, and the outer peripheral flatness is calculated for each section of the divided outer peripheral region A double-sided polishing apparatus characterized by the above.

4. In the double-sided polishing apparatus according to claim 3, When a section adjacent to the in-plane region is defined as the innermost section and a section radially outside the innermost section is defined as the outer section, the flatness of the innermost section is obtained by performing a continuity process on the inclination values of the approximate straight lines set for the thickness data group of the innermost section based on the in-plane flatness, the flatness of the outer section is obtained by performing a continuity process on the inclination values of the approximate straight lines set for the thickness data group of the outer section based on the flatness of the section adjacent to the inner side in the radial direction of the outer section A double-sided polishing apparatus characterized by the above.

5. In a double-sided polishing method in which a disk-shaped workpiece is sandwiched between a lower polishing plate and an upper polishing plate disposed opposite to the lower polishing plate, and the lower polishing plate, the upper polishing plate, and the workpiece are relatively moved while applying a load to the workpiece to polish the front and back surfaces of the workpiece, a first learning device that learns the correlation between the polishing conditions of the main polishing for polishing the workpiece while maintaining the load at a target load, the outer peripheral flatness that is the flatness of an outer peripheral region set in a predetermined range from the outer peripheral end of the workpiece toward the inner side in the radial direction, and the in-plane flatness that is the flatness of the in-plane region in the range from the outer peripheral region of the workpiece to the workpiece center; a second learning device that learns the correlation between the polishing conditions of the final polishing for polishing the workpiece while gradually reducing the load, the degree of change in the in-plane flatness at the time of stopping the main polishing, the degree of change in the outer peripheral flatness at the time of stopping the main polishing, and the amount of change in the outer peripheral flatness during the final polishing; and using a step of setting a target value for the outer peripheral flatness and a target value for the in-plane flatness; After setting the target values for the outer peripheral flatness and the in-plane flatness, a step of performing initial polishing for polishing the workpiece while gradually increasing the load; After the initial polishing is completed, a step of starting the main polishing; During the execution of the main polishing, based on the polishing conditions of the main polishing set at the correction time and the estimated value of the in-plane flatness obtained by inputting the outer peripheral flatness at the correction time into the first learning device, a correction target value of the in-plane flatness is obtained. The polishing conditions of the main polishing set at the correction time are corrected by the polishing conditions of the main polishing obtained by inputting the correction target value of the in-plane flatness and the target value of the outer peripheral flatness into the first learning device. During the execution of the main polishing, the change amount of the outer peripheral flatness during the finish polishing is obtained by inputting the polishing conditions of the finish polishing, the degree of change of the in-plane flatness at the calculation time, and the degree of change of the outer peripheral flatness at the calculation time into the second learning device. A stop condition for the main polishing is set based on the change amount of the outer peripheral flatness and the target value of the outer peripheral flatness. After the stop condition of the main polishing is satisfied, the finish polishing is executed. A double-sided polishing method characterized by comprising the above.

Images