Teaching method and semiconductor manufacturing apparatus

Abstract

To provide a technique capable of accurately acquiring the position of a circular measuring object.SOLUTION: The teaching method teaches a position of a circular measurement target in a semiconductor manufacturing apparatus including a transfer device having a fork capable of transferring a substrate and a control device configured to control an operation of the transfer device. A step (A) of measuring an edge portion of a circular measurement target a plurality of times by a sensor provided on a fork, a step (B) of creating a measurement virtual circle based on a plurality of measurement points measured by the sensor, a step (C) of calculating an error between the measurement virtual circle and the plurality of measurement points, a step (D) of creating an extracted virtual circle by excluding an outlier measurement point among the plurality of measurement points based on the error, and a step (E) of comparing the extracted virtual circle with the excluded outlier measurement point, and adding the measurement point of the outlier that can be added to the extracted virtual circle and calculating the center position of the final circle.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present disclosure relates to teaching methods and semiconductor manufacturing equipment. [Background technology]

[0002] Patent Document 1 discloses a center detection method that detects the outer edge of a wafer using an imaging unit and calculates the center position of the wafer based on the coordinates of the outer edge. In this center detection method, an approximate circle is calculated from the coordinates of three or more points on the outer edge, and coordinates that exceed a threshold value are excluded from the calculated approximate circle as erroneous recognition, and the approximate circle is recalculated to calculate the center position of the wafer.

[0003] Furthermore, Cited Document 2 discloses a technique for detecting the outer edge of a substrate (wafer) being transported and calculating the center of the substrate using a pair of sensors installed in a substrate transport chamber of a plasma processing apparatus, which is a semiconductor manufacturing system. In this semiconductor manufacturing system, the center of the substrate is calculated by excluding the outer edge (edge ​​intersection) that exists in an abnormal position. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 7437958 [Patent Document 2] Patent No. 6640321 Summary of the Invention [Problem to be solved by the invention]

[0005] The present disclosure provides a technique that can accurately acquire the position of a circular measurement target. [Means for solving the problem]

[0006] According to one aspect of the present disclosure, there is provided a teaching method for teaching the position of a circular measurement target in a semiconductor manufacturing apparatus including a transport device having a fork capable of transporting a substrate and a control device that controls the operation of the transport device, the teaching method comprising: (A) a step of measuring an edge of the circular measurement target multiple times using a sensor provided on the fork; (B) a step of creating a measurement virtual circle based on the multiple measurement points measured by the sensor; (C) a step of calculating an error between the measurement virtual circle and the multiple measurement points; (D) a step of creating an extracted virtual circle by excluding outlier measurement points from the multiple measurement points based on the error; and (E) a step of comparing the extracted virtual circle with the excluded outlier measurement points, and adding the outlier measurement points that can be added to the extracted virtual circle, thereby calculating the final center position of the circle. [Effects of the Invention]

[0007] According to one aspect, the position of a circular measurement target can be acquired with high accuracy. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a schematic plan view showing an example of the overall configuration of a substrate processing system according to an embodiment; [Figure 2] FIG. 2 is an enlarged plan view showing the fork of the vacuum transfer device and the substrate. [Figure 3] FIG. 2 is a cross-sectional view schematically illustrating an example of a processing module. [Figure 4] FIG. 10 is an enlarged cross-sectional view showing a state in which a ring and a substrate are placed on a substrate support part. [Figure 5] Fig. 5(A) is a side view illustrating the principle of teaching the transfer position of a vacuum transfer device, Fig. 5(B) is a plan view illustrating the principle of teaching the transfer position of a vacuum transfer device, and Fig. 5(C) is a plan view illustrating an example of detecting multiple outer edges in the teaching method according to the embodiment. [Figure 6] Fig. 6(A) is a cross-sectional view showing the detection of the inner peripheral surface of the step of the ring by the sensor, and Fig. 6(B) is a cross-sectional view showing the detection of the sensor when deposits adhere to the ring. [Figure 7] FIG. 1 is a flowchart illustrating a teaching method according to an embodiment. [Figure 8] Fig. 8(A) is a first explanatory diagram for explaining the processing of the teaching method, Fig. 8(B) is a second explanatory diagram for explaining the processing of the teaching method, and Fig. 8(C) is a third explanatory diagram for explaining the processing of the teaching method. [Figure 9] Fig. 9(A) is a first explanatory diagram for explaining the processing of the teaching method, Fig. 9(B) is a second explanatory diagram for explaining the processing of the teaching method, and Fig. 9(C) is a third explanatory diagram for explaining the processing of the teaching method. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In the drawings, the same components are denoted by the same reference numerals, and redundant explanations may be omitted.

[0010] 1, a semiconductor manufacturing apparatus according to an embodiment of the present disclosure constitutes a multi-chamber type substrate processing system 1 including a plurality of (six) processing modules 10 for processing substrates W. Note that the semiconductor manufacturing apparatus is not limited to this substrate processing system 1, and may be, for example, an apparatus for inspecting substrates W, an apparatus for measuring the state (film thickness, etc.) of substrates W, an apparatus for only transporting substrates W, an apparatus for keeping substrates W waiting, or the like.

[0011] Each processing module 10 of the substrate processing system 1 performs substrate processing such as film formation processing, etching processing, modification processing, cleaning processing, bonding processing, peeling processing, ashing processing, etc. on the substrate W transferred therein. In addition to each processing module 10, the substrate processing system 1 also includes a vacuum transfer module 20, a plurality of load lock modules 30, an atmospheric transfer module 40, a load port 50, a control device 90, etc.

[0012] Each processing module 10 carries substrates W in and out of the vacuum transfer module 20, and performs substrate processing on the carried-in substrates W. Of course, the number of processing modules 10 provided in the substrate processing system 1 is not particularly limited. Furthermore, the multiple processing modules 10 may perform the same processing, or some or all of them may perform different processing. The substrate processing system 1 may be configured to perform plasma processing in some or all of the processing modules 10.

[0013] Each processing module 10 has a processing vessel 11 that accommodates a substrate W, and a substrate support part 12 on which the substrate W is placed inside the processing vessel 11. The substrate support part 12 includes a lifter part 19 (see FIG. 3) that raises and lowers the substrate W, and receives and delivers the substrate W in cooperation with a vacuum transfer device 22 of a vacuum transfer module 20, which will be described later.

[0014] Each processing module 10 is connected to a vacuum transfer module 20 via a connection unit 15. The connection unit 15 has therein a gate valve 15g (see FIG. 3) that opens and closes the loading / unloading port 11p of the processing vessel 11. Each processing module 10 can transfer a substrate W to the processing vessel 11 by opening the gate valve 15g, and can depressurize the inside of the processing vessel 11 to an appropriate vacuum atmosphere by closing the gate valve 15g.

[0015] The vacuum transfer module 20 of the substrate processing system 1 includes a transfer container 21 connected to each processing module 10 and each load lock module 30, and a vacuum transfer device 22 provided in the transfer container 21 for transferring the substrate W. The vacuum transfer module 20 may also include a plurality of transfer regions (or transfer containers 21) each having the vacuum transfer device 22, and a pass region connecting the transfer regions, and may be configured to transfer the substrate W from one transfer region to another transfer region via the pass region.

[0016] The transfer container 21 has a rectangular shape in a plan view and includes a transfer space 21s that is airtightly closed from the outside. The transfer space 21s is depressurized to a vacuum atmosphere by a vacuum suction device (not shown). In the substrate processing system 1 according to the embodiment, three processing modules 10 are connected to each of a pair of long sides of the transfer container 21. In addition, in the substrate processing system 1, two load lock modules 30 are connected to one short side of the transfer container 21.

[0017] The vacuum transfer device 22 moves within the transfer space 21s to transfer the substrate W under the control of the control device 90. For example, the vacuum transfer device 22 transfers the substrate W from one load lock module 30 to the target processing module 10. The vacuum transfer device 22 also transfers the substrate W from one processing module 10 to the target load lock module 30. The vacuum transfer device 22 may also transfer the substrate W between two processing modules 10. Note that the transported object transported by the vacuum transfer device 22 is not limited to the substrate W. For example, when a ring R, which will be described later and which is applied to the processing module 10, is transported via the vacuum transfer module 20 and set in the processing module 10, the transported object transported by the vacuum transfer device 22 may include the ring R.

[0018] The vacuum transfer device 22 has a base 221 that is movable in the longitudinal direction of the transfer container 21, a plurality of arms 222 that can freely rotate, extend, and lower relative to the base 221, and a fork (end effector) 223 provided on the distal arm 222. Note that, although Fig. 1 illustrates the vacuum transfer device 22 having two forks 223, the present invention is not limited to this, and the vacuum transfer device 22 may have one or three or more forks 223.

[0019] As shown in FIG. 2, the vacuum transfer device 22 supports the substrate W on the upper surface of the fork 223, and transfers the substrate W by appropriately operating the base 221 and each arm 222 shown in FIG.

[0020] The fork 223 includes a base plate 224 connected to the distal arm 222, and a pair of support plates 225 that extend in two directions from the base plate 224. The base plate 224 and the pair of support plates 225 are integrally molded and connected horizontally, forming a U-shape in plan view. The pair of support plates 225 extend parallel to each other and have the same length. A recessed space 223s surrounded by the base plate 224 and the pair of support plates 225 opens at the tip ends of the pair of support plates 225.

[0021] The fork 223 also includes a plurality of pads 226 on the upper surfaces of the base plate portion 224 and the support plate portion 225. For example, the plurality of pads 226 are provided at a widthwise intermediate position of the base plate portion 224 and at the tip ends of the pair of support plate portions 225, thereby directly supporting three portions of the substrate W. The plurality of pads 226 may be formed of a material such as an elastomer having appropriate frictional force and elasticity. Furthermore, the fork 223 may be configured to hold the substrate W by including a holding means such as a suction mechanism, an electrostatic suction mechanism, or a mechanical locking mechanism that utilizes (or replaces) the plurality of pads 226.

[0022] When the vacuum transfer device 22 supports the substrate W with the fork 223, the fork 223 is moved so that the reference position 223o coincides with the transfer position instructed by the control device 90. The transfer position is, for example, the center position Wo of the substrate W, the center position Ro of the ring R (see FIG. 4), the center position 12o of the substrate support part 12 (see FIG. 4), etc.

[0023] The fork 223 is provided with a sensor 227 for detecting the substrate W, which is a transported object, on the surface (lower surface) opposite to the surface that supports the substrate W. The sensor 227 according to the embodiment has a plurality of (two in FIG. 2) detectors 227a, 227b. Each of the detectors 227a, 227b is provided near the extending ends (tips of the fork 223) of the pair of support plate portions 225, and detects an object present (facing) below the fork 223 in the vertical direction. The sensor 227 is also communicatively connected to the control device 90, performs detection under the control of the control device 90, and transmits the detection information to the control device 90.

[0024] For example, the sensor 227 may be a displacement sensor that optically measures the distance from the fork 223 to an object. In this case, each of the detectors 227a, 227b of the sensor 227 has a light-emitting unit and a light-receiving unit, and measures the distance to the object based on the light intensity or wavelength of the detection light emitted by the light-emitting unit and reflected by the object. Alternatively, each of the detectors 227a, 227b may be configured to detect changes in the height of the object by switching on and off. For example, as the fork 223 moves above the substrate W and the sensor 227 continues to perform detection, the sensor 227 detects changes in height at the outer edge (periphery, edge) of the substrate W by switching on and off. The control device 90 acquires the position of the outer edge of the substrate W as detection information from the sensor 227, and can calculate the center position Wo of the substrate W from the positions of the multiple outer edges using a method described below.

[0025] The sensor 227 is not particularly limited in type as long as it can acquire the center position of the circular measurement target. Other examples of the sensor 227 include an on-off sensor that detects the outer edge of the substrate W based on the transmission or blocking of detection light, and a capacitance sensor that detects a change in capacitance when the substrate W passes above. Alternatively, the sensor 227 may be an infrared sensor, an ultrasonic sensor, a radar, a camera, or the like. The fork 223 is not particularly limited in terms of the position of the sensor 227 or the number of detectors, and may have, for example, three or more detectors.

[0026] 1, the two load lock modules 30 of the substrate processing system 1 are provided between the vacuum transfer module 20 and the atmospheric transfer module 40, and switch the interior between an atmospheric atmosphere and a vacuum atmosphere. Specifically, each load lock module 30 includes a container 31 for accommodating a substrate W, and a mounting table 32 on which the substrate W is placed inside the container 31. For example, the mounting table 32 includes grooves (not shown) into which the forks 223 of the vacuum transfer device 22 and the forks 423 of the atmospheric transfer device 42 described below can enter, and the substrate W is received and transferred by the forward / backward movement and vertical movement of the forks 223, 423. The mounting table 32 may be configured to include a lifter unit similar to the substrate support unit 12 of the processing module 10.

[0027] Each load lock module 30 has a connection part 33 on the vacuum transfer module 20 side and a connection part 35 on the atmospheric transfer module 40 side. The connection parts 33, 35 have gate valves (not shown) inside that open and close the openings of the containers 31. Each load lock module 30 communicates with the vacuum transfer module 20 when the gate valve of the connection part 33 is opened in a vacuum atmosphere state. Each load lock module 30 also communicates with the atmospheric transfer module 40 when the gate valve of the connection part 35 is opened in an atmospheric atmosphere state.

[0028] The atmospheric transfer module 40 of the substrate processing system 1 transfers the substrate W while maintaining the interior at an atmospheric atmosphere. The atmospheric transfer module 40 includes a transfer container 41 connected to each load lock module 30, and an atmospheric transfer device 42 provided in the transfer container 41 for transferring the substrate W. The atmospheric transfer module 40 may be configured so that clean air flows down into the transfer container 41. The atmospheric transfer module 40 also includes an aligner 43 on its side for aligning the substrate W.

[0029] Furthermore, a plurality of load ports 50 are provided on one side of the atmospheric transfer module 40. A carrier C containing a substrate W or an empty carrier C is set in each load port 50. For example, a FOUP (Front Opening Unified Pod) or the like can be used as the carrier C. Furthermore, a carrier C containing a ring R (focus ring, edge ring, etc.), which is an example of a transported object, may be set in each load port 50. The ring R is arranged around the substrate W on the substrate support part 12 of the processing module 10 (see FIG. 3).

[0030] Similar to the vacuum transfer device 22, the atmospheric transfer device 42 has a base 421 that is movable in the longitudinal direction of the transfer container 41, a plurality of arms 422 that are rotatable, extendable, and movable up and down relative to the base 421, and a fork (end effector) 423 provided on the arm 422 at the distal end. The atmospheric transfer device 42 supports the substrate W on the upper surfaces of the forks 423, and transfers the substrate W by appropriately operating the base 421 and each arm 422. Note that, although FIG. 1 illustrates an atmospheric transfer device 42 that has two forks 423, the present invention is not limited to this, and the atmospheric transfer device 42 may have a configuration that includes one or three or more forks 423.

[0031] The atmospheric transfer device 42 loads and unloads the substrate W between each load lock module 30 and the atmospheric transfer module 40 in response to the opening and closing of the gate valves of each connection part 35. The atmospheric transfer device 42 also loads and unloads the substrate W between the aligner 43 and the atmospheric transfer module 40. The atmospheric transfer device 42 also loads and unloads the substrate W between each carrier C attached to each load port 50 and the atmospheric transfer module 40.

[0032] The control device 90 is a computer having a processor 91, a memory 92, and an input / output interface and a communication interface (not shown). The processor 91 is one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a circuit made up of a plurality of discrete semiconductors, etc. The memory 92 includes a main memory device and an auxiliary memory device, and stores a program for controlling the substrate processing system 1.

[0033] For example, the control device 90 transfers an unprocessed substrate W in a carrier C set in the load port 50 to the aligner 43 using the atmospheric transfer device 42 for alignment, and then transfers the substrate W in the aligner 43 to one of the load lock modules 30. Then, the control device 90 depressurizes the load lock module 30 accommodating the substrate W, removes the substrate W using the vacuum transfer device 22, and transfers the substrate W into the target processing module 10 via the vacuum transfer module 20. Thereafter, the control device 90 processes the substrate in each processing module 10 into which the substrate W has been carried. After the substrate processing, the control device 90 transfers the substrate W from the target processing module 10 to a carrier C for accommodating the processed substrate W by reversing the above procedure.

[0034] Next, an example of a processing module 10 that is applied to the above-described substrate processing system 1 will be described with reference to Fig. 3. The processing module 10 performs substrate processing on the substrate W under the control of the control device 90 of the substrate processing system 1. However, the processing module 10 may also have a dedicated control circuit and perform substrate processing under the control of the dedicated control circuit that receives commands from the control device 90.

[0035] Specifically, the processing module 10 includes the processing vessel 11 and the substrate support unit 12 described above, as well as a gas supply unit 16 , a gas exhaust unit 17 , a power supply unit 18 , and a lifter unit 19 .

[0036] The processing vessel 11 has an internal processing space 11s for accommodating a substrate W. A loading / unloading port 11p is provided in a sidewall of the processing vessel 11. The substrate W is transferred between the processing space 11s and the outside of the processing vessel 11 through the loading / unloading port 11p. The loading / unloading port 11p is opened and closed by a gate valve 15g of the connection part 15.

[0037] The substrate support 12 is disposed in a lower region of a processing space 11s in the processing vessel 11. The substrate support 12 supports a substrate W in the processing space 11s. The substrate support 12 includes a lower electrode 121, an electrostatic chuck 122, a ring R (such as a focus ring or an edge ring), and an insulating member 125.

[0038] The electrostatic chuck 122 is stacked on the upper side of the lower electrode 121. The electrostatic chuck 122 has an upper surface including a substrate support surface 122a and a ring support surface 122b. The electrostatic chuck 122 supports the substrate W on the substrate support surface 122a. The electrostatic chuck 122 supports the ring R on the ring support surface 122b. The electrostatic chuck 122 has a first chucking electrode 122d and a second chucking electrode 112e. The first chucking electrode 122d and the second chucking electrode 112e are embedded in the dielectric of the electrostatic chuck 122. The first chucking electrode 122d is located below the substrate support surface 122a. The electrostatic chuck 122 attracts and holds the substrate W on the substrate support surface 122a by applying a voltage to the first chucking electrode 122d. The second chucking electrode 112e is located below the ring support surface 122b. The electrostatic chuck 122 attracts and holds the ring R on the ring support surface 122b by applying a voltage to the second attracting electrode 112e.

[0039] The ring R is a focus ring or edge ring formed in an annular shape and placed on the ring support surface 122b so as to surround the substrate W in order to improve the uniformity of plasma processing on the substrate W. The ring R is made of a conductive material such as silicon (Si) or silicon carbide (SiC). The ring R may also be a cover ring made of an insulating material such as quartz. The ring R may be made of a single member or a combination of multiple members.

[0040] An insulating member 125 is disposed to surround the lower electrode 121 and the electrostatic chuck 122. The insulating member 125 is fixed to the bottom of the processing chamber 11 and supports the lower electrode 121.

[0041] The processing vessel 11 also includes a shower head 13 above the substrate support 12 across the processing space 11s. The shower head 13, together with an insulator 14, forms a ceiling of the processing vessel 11. The shower head 13 supplies one or more types of processing gas from a gas supply unit 16 to the processing space 11s. The shower head 13 includes a top plate 131 and a support 132. The top plate 131 is provided with a plurality of gas inlets 131a. Each of the plurality of gas inlets 131a penetrates the top plate 131 in the thickness direction (vertical direction). Meanwhile, a gas diffusion chamber 132a is provided inside the support 132, to which the plurality of gas inlets 131a communicate. The support 132 is also provided with a gas supply port 132b. The shower head 13 supplies one or more types of processing gas from a gas supply port 132b through a gas diffusion chamber 132a, a plurality of gas inlets 131a, and a plurality of gas inlets 131b to the processing space 11s. The shower head 13 may be connected to a power supply unit 18 to form an upper electrode capable of generating plasma in the processing gas.

[0042] The gas supply unit 16 has one or more gas sources 161 and a number of flow rate controllers 162 corresponding to the number of gas sources 161. The gas supply unit 16 supplies one or more types of process gases from the gas sources 161 to the gas supply port 132b via the flow rate controllers 162. The flow rate controllers 162 may be, for example, mass flow controllers or pressure-controlled flow rate controllers.

[0043] The gas exhaust unit 17 is connected to a gas exhaust port 11e provided at the bottom of the processing vessel 11, for example. The gas exhaust unit 17 includes, for example, a pressure adjustment valve and a vacuum pump. The pressure adjustment valve adjusts the pressure in the processing space 11s. The vacuum pump may include a turbomolecular pump, a dry pump, or a combination thereof.

[0044] The power supply unit 18 generates plasma in the processing space 11s by supplying power to the lower electrode 121. For example, the power supply unit 18 includes two RF power supplies (a first RF power supply 181a and a second RF power supply 181b) and two matchers (a first matcher 182a and a second matcher 182b). The first RF power supply 181a supplies a first RF power to the lower electrode 121 via the first matcher 182a. The frequency of the first RF power may be, for example, 13 MHz to 150 MHz. The second RF power supply 181b supplies a second RF power to the lower electrode 121 via the second matcher 182b. The frequency of the second RF power may be, for example, 400 kHz to 13.56 MHz. A DC power supply may be used instead of the second RF power supply 181b. The power supply unit 18 may also be configured to supply a bias power to the lower electrode 121 that attracts active species in the plasma.

[0045] The lifter unit 19 includes a first lifter 191 and a second lifter 192. The first lifter 191 has a plurality of support pins 191a and an actuator 191b. The plurality of support pins 191a are raised and lowered by operation of the actuator 191b and protrude from the substrate support surface 122a of the electrostatic chuck 122, thereby contacting the lower surface of the substrate W and supporting the substrate W. The second lifter 192 has a plurality of support pins 192a and an actuator 192b. The plurality of support pins 192a are raised and lowered by operation of the actuator 192b and protrude from the ring support surface 122b of the electrostatic chuck 122, thereby contacting the lower surface of the ring R and supporting the ring R.

[0046] 4, the upper surface of the ring R placed on the substrate support 12 has, for example, an outer annular surface Rout that is located radially outward from the outer edge of the substrate W, and an inner annular surface Rin that is located inside the outer annular surface Rout and faces the underside of the substrate W. The inner annular surface Rin is lower than the outer annular surface Rout. In other words, the upper surface of the ring R has a stepped structure with an annular stepped inner peripheral surface RS between the outer annular surface Rout and the inner annular surface Rin.

[0047] For example, the inner annular surface Rin faces, without contacting, the lower surface of the substrate W placed on the substrate support surface 122a. Therefore, the portion having the inner annular surface Rin is pre-adjusted to have a thickness that is lower than the substrate support surface 122a when the ring R is placed on the ring support surface 122b. On the other hand, the portion having the outer annular surface Rout is pre-adjusted to have a thickness that is approximately the same as the upper surface (front surface) of the substrate W placed on the substrate support surface 122a. Furthermore, the step inner peripheral surface RS has an inner diameter that is slightly larger than the outer diameter of the substrate W. Therefore, when the substrate W is placed, etc., a small gap is generated between the outer edge of the substrate W and the step inner peripheral surface RS, which suppresses rubbing between the outer edge of the substrate W and the ring R.

[0048] In the above-described substrate processing system 1, when the vacuum transfer device 22 transfers the substrate W to the processing module 10, the substrate W is placed on the substrate supporting surface 122a so that the center position 12o of the substrate supporting surface 122a coincides with the center position Wo of the substrate W. Note that, for ease of understanding, Fig. 4 shows a state in which the center position Wo of the substrate W is intentionally shifted from the center position 12o of the substrate supporting surface 122a of the substrate support part 12.

[0049] If a misalignment occurs between the center position 12o of the substrate support surface 122a and the center position Wo of the substrate W, it may affect the in-plane uniformity of the plasma processing. For this reason, the substrate processing system 1 needs to accurately teach the control device 90 about the transfer position within the processing module 10 when installing the system, performing maintenance, replacing the ring R, etc.

[0050] Next, a method for teaching the transfer position of the vacuum transfer device 22 will be described with reference to FIGS. 5A and 5B. As shown in FIGS. 5A and 5B, in this teaching method, the sensor 227 of the vacuum transfer device 22 detects multiple edges of a circular measurement target, and the center position of the measurement target can be calculated from the positions of the multiple detected measurement points. Below, a representative teaching method for teaching the transfer position for transporting the substrate W to the substrate support unit 12 will be described. In the case of the electrostatic chuck 122 (substrate support unit 12) on which the ring R is mounted, for example, the step inner peripheral surface RS of the ring R is detected as the edge of the circular measurement target. Note that the circular measurement target detected by the sensor 227 is not limited to the step inner peripheral surface RS of the ring R, but may also be the outer edge of the outer annular surface Rout of the ring R or the inner edge of the inner annular surface Rin of the ring R. Alternatively, the circular measurement target may be the outer edge of the substrate support surface 122a of the electrostatic chuck 122 or the outer edge of the ring support surface 122b. Furthermore, when the fork 223 transports the substrate W or the ring R, it is necessary to recognize the center position Wo of the substrate W or the center position Ro of the ring R and hold the substrate W or the ring R with the fork 223. Therefore, the circular measurement object may be the substrate W or the ring R.

[0051] The position of the stepped inner peripheral surface RS of the ring R serves as an index for deriving the center position Ro of the ring R, which is used when placing the substrate W inside the ring R. That is, when placing the substrate W on the electrostatic chuck 122 on which the ring R is placed, the substrate W can be placed on the electrostatic chuck 122 with high precision by aligning the derived center position Ro of the ring R with the center position Wo of the substrate W.

[0052] Specifically, the control device 90 detects (scans) the substrate W using the sensor 227 while sliding the vacuum transfer device 22 horizontally and linearly above the vertically upper side of the ring R. At this time, the linear movement of the vacuum transfer device 22 is controlled so that the reference position 223o of the fork 223 passes through a preset design position of the substrate support part 12.

[0053] The sensor 227, provided near the tip of each support plate portion 225 of the vacuum transfer device 22, detects the step inner peripheral surface RS of the ring R as it passes above the ring R. As shown in FIG. 6A, each of the detectors 227a and 227b of the sensor 227 transmits the timing at which the amount of received light changes to the control device 90 as it passes above the ring R. Furthermore, the control device 90 recognizes the position of the fork 223, in other words, the position (three-dimensional coordinates) of each of the detectors 227a and 227b, based on the movement of each arm 222 of the vacuum transfer device 22 during scanning by the vacuum transfer device 22. The control device 90 can recognize four measurement points Rd (see white stars in FIG. 5B) on the step inner peripheral surface RS of the ring R by linking the timing of the detection information from the sensor 227 with the recognized positions of each of the detectors 227a and 227b. The substrate processing system 1 may measure the distance to the ring R using the sensor 227, and obtain each measurement point Rd on the step inner circumferential surface RS based on the change in this distance.

[0054] Here, the control device 90 can use the positions of the four detected measurement points Rd to calculate the center position Ro of the ring R (see the black stars in FIG. 5(B)). For example, the control device 90 can create two sets of right-angled triangles connecting two of the four detected measurement points Rd on the far side and one on the near side, and calculate the center position Ro of the ring R as the average of the center coordinates of the two sets of right-angled triangles.

[0055] However, when using the four measurement points Rd detected by the sensor 227, the position of the measurement points Rd is easily affected by changes in the surface condition of the ring R, which is the measurement target. Specifically, as shown in FIG. 6(B), the surface condition of the ring R changes (for example, the surface becomes uneven, different from when it was new) due to the adhesion of deposits and wear due to etching during substrate processing in the processing module 10. Note that deposits are indicated by black dots in FIG. 6(B). Due to this change in the surface condition, for example, the detection information of the sensor 227 repeats minute amplitudes. When the control device 90 receives detection information with such repeating amplitudes from the sensor 227, the possibility of erroneously detecting the position of the measurement points Rd increases.

[0056] Therefore, in the substrate processing system 1 according to the embodiment, the movement of the fork 223 and the measurement of the step inner peripheral surface RS of the ring R by the sensor 227 are performed multiple times (for example, three times). Then, the control device 90 is configured to create a least-squares circle by collectively excluding measurement points Rd with low reliability from the four measurement points Rd obtained each time, and calculate the coordinates of the center position of the least-squares circle. In other words, when the measurement of the step inner peripheral surface RS is performed three times, a total of 12 measurement points Rd are obtained.

[0057] Furthermore, as shown in FIG. 5C , when performing multiple measurements, it is preferable to shift the fork 223 laterally, perpendicular to the forward / backward direction, and then move the fork 223 forward / backward to change the position of the step inner peripheral surface RS of the ring R measured by the sensor 227. For example, if the reference position for the fork 223 when moving forward / backward is a line along a white circle, the fork 223 is moved forward / backward along a line along a white triangle shifted by +α in the horizontal direction from the reference position, and along a line along a white square shifted by -α in the horizontal direction from the reference position. As a result, the control device 90 obtains measurement points Rd at 12 different positions during three measurements of the step inner peripheral surface RS. The control device 90 creates a least-squares circle using these multiple measurement points Rd and improves the accuracy of the least-squares circle by omitting measurement points Rd that contain errors.

[0058] Hereinafter, a teaching method for calculating the center position Ro of the stepped inner peripheral surface RS of the ring R in the vacuum transfer device 22 will be specifically described with reference to the flowchart of Fig. 7. In the teaching method, the control device 90 sequentially performs the processes of steps S101 to S110 shown in Fig. 7.

[0059] In the teaching method, the control device 90 first measures the step inner peripheral surface RS of the ring R multiple times (three times in this embodiment) using the sensor 227 of the fork 223 (step S101). As described above, in the multiple measurements, the position of the fork 223 is shifted laterally for each measurement, and then the fork 223 is moved forward and backward, thereby obtaining different positions on the step inner peripheral surface RS as measurement points Rd (see FIG. 5(C)). The control device 90 then proceeds to a process of calculating the center position Ro from each of the measurement points Rd obtained internally.

[0060] The control device 90 sets a threshold value for comparison with the mean square error (described later) (step S102). The threshold value may be set in advance through simulation, experiment, or the like to an appropriate value that allows for distinguishing measurement points Rd of the ring R to be measured. For example, the threshold value for the mean square error may be set in the range of approximately 0.05 to 0.2. In this embodiment, the threshold value is set to 0.1. Note that this threshold value setting may be performed before step S105, when the threshold value is actually used.

[0061] Then, the control device 90 creates measured least-squares circles I1 to I3 (measured virtual circles) for each combination of measurement points obtained from multiple measurements (step S103). That is, if three measurements are taken by shifting the position laterally using the fork 223, three measured least-squares circles I1 to I3 are created by each measurement point Rd for each of the three measurements, as shown in Fig. 8(A). The least-squares circle is the circle for which the sum of the difference between the square of the distance from each measurement point Rd to the center of the circle and the square of the circle's radius is the smallest.

[0062] Next, the control device 90 calculates the mean square error between the created measured least squares circles I1-I3 and the measurement point Rd used (step S104). The mean square error is a measure of the quality of the measured least squares circle (estimated value) and approaches zero as the error of the measurement point Rd decreases. In other words, a measured least squares circle with a large mean square error can be considered to include a measurement point Rd with a large positional deviation. For example, FIG. 8(B) shows an example in which the mean square error of the measured least squares circle I1 for +α is 0.3345, the mean square error of the measured least squares circle I2 for the reference position is 0.0005, and the mean square error of the measured least squares circle I3 for -α is 0.0046.

[0063] The control device 90 then compares the calculated mean squared errors of each of the measured least squares circles I1-I3 with a preset threshold, and creates a least squares circle (hereinafter referred to as an extracted least squares circle IE) again at the measurement points Rd that form a circle that is equal to or less than the threshold (step S105). As set in step S102, the threshold for comparison is, for example, 0.1. Therefore, in the example of the measured least squares circles I1-I3 in FIG. 8(B), the measured least squares circles I2 and I3 are extracted as circles whose mean squared errors are equal to or less than the threshold.

[0064] As a result, as shown in Figure 8(C), the control device 90 creates the extracted least-squares circle IE using the measurement points Rd on the measurement least-squares circles I2 and I3. Note that in Figure 8(C), the measurement points Rd used in calculating the extracted least-squares circle IE are shown in black, and unused measurement points Rd are shown in white. In other words, the unused measurement points Rd at this point are outliers that were excluded from the extracted least-squares circle IE in the initial stage.

[0065] However, if fewer than half of the measured least-squares circles I1-I3 have a mean squared error below the threshold, or if the mean squared error of the created extracted least-squares circle I.E. is above the threshold, the control device 90 may return to step S101 and redo the measurement of the fork 223 using the sensor 227 or increase the number of measurements. This is because if half of the measured least-squares circles I1-I3 exceed the threshold, or if the mean squared error of the created extracted least-squares circle I.E. is above the threshold, the measurement results of the sensor 227 contain a large amount of error, raising concerns about abnormal operation of the fork 223 or other measurement abnormalities. By redoing the measurement, the control device 90 can improve the accuracy of the calculated extracted least-squares circle. Alternatively, if the mean squared error of more than half of the measured least-squares circles I1-I3 is not below the threshold even after redoing the measurement, or if the mean squared error of the created extracted least-squares circle I.E. is above the threshold, the control device 90 may notify the user of an error.

[0066] Thereafter, the control device 90 calculates the squared error of each unused measurement point Rd for the created extracted least-squares circle IE (step S106). For example, as shown in FIG. 9(A), the control device 90 calculates the squared error of each measurement point Rd indicated by an open triangle, and then ranks the points in ascending order of squared error. Note that in FIG. 9(A), the decimal point indicates the calculated squared error value, and the number in parentheses next to it indicates the rank.

[0067] Furthermore, the control device 90 adds the ranked measurement points Rd with the smallest squared errors to each of the measurement points Rd constituting the current extracted least-squares circle IE, and recreates the extracted least-squares circle IE (step S107). For example, in FIG. 9A, the control device 90 adds the measurement points Rd ranked (1) to (4) starting from the first one (the measurement point Rd with the smallest squared error). FIG. 9B shows an example in which the extracted least-squares circle IE is recreated by adding the first measurement point Rd of the measurement points Rd indicated by the open triangles.

[0068] Then, after recreating the extracted least squares circle IE, the control device 90 calculates its mean square error (step S108). This calculation of the mean square error is intended to verify which measurement points Rd can be added among the multiple outliers (measurement points Rd) that were once excluded.

[0069] Therefore, the control device 90 compares the calculated mean square error with a set threshold (e.g., 0.1) to determine whether the mean square error exceeds the threshold (step S109). If the mean square error is equal to or less than the threshold (step S109: NO), the control device 90 uses the added measurement points Rd as is, returns to step S107, and repeats step S107 and subsequent steps. In this way, by repeatedly adding (restoring) outlier measurement points Rd in ascending order of square error and calculating the extracted least squares circle until the mean square error exceeds the threshold, the accuracy of the extracted least squares circle IE can be improved. For example, the control device 90 can improve the accuracy of circle calculation by selecting appropriate measurement points Rd even from measurement points Rd where deposits or wear have occurred.

[0070] On the other hand, if the mean square error exceeds the threshold (step S109: YES), adding any more outlier measurement points Rd will not improve the accuracy of the extracted least squares circle. Therefore, the control device 90 does not add any more measurement points Rd, but instead adopts the extracted least squares circle created previously. The control device 90 then recognizes the extracted least squares circle IE adopted in the determination of step S109 (the extracted least squares circle including the outlier measurement points Rd that are equal to or smaller than the threshold) as the final step inner peripheral surface RS of the ring R.

[0071] As a result, the control device 90 calculates the center coordinates from the extracted least-squares circle IE used in step S109, applies these center coordinates as the center position Ro of the ring R, and calculates the radius from the center coordinates to the extracted least-squares circle IE (step S110). Figure 9(C) shows an example in which the center position Ro is calculated when only one of the measurement points Rd indicated by the open triangles that were outliers is added and the other measurement points Rd are excluded. The center coordinates of the extracted least-squares circle IE can be easily calculated using a well-known method.

[0072] As described above, in the teaching method, an outlier is first set for each measurement point Rd obtained through multiple measurements, and then the outliers with the smallest error are added sequentially to the extracted least-squares circle I. In other words, the teaching method can maximize the use of the measurement points Rd while simultaneously eliminating measurement points Rd with outliers, thereby improving the accuracy of calculating the center position of the measurement object. The control device 90 then sets the center position Ro of the ring R obtained by the teaching method described above as the transfer position of the vacuum transfer device 22. This enables the control device 90 to accurately align the center position Wo of the substrate W supported by the fork 223 with the center position Ro of the ring R (the center position 12o of the substrate support 12).

[0073] The teaching method and semiconductor manufacturing apparatus of the present disclosure are not limited to the above-described embodiment, and various modifications are possible. For example, the above-described embodiment describes a teaching method for the vacuum transfer device 22. However, the teaching method is not limited to this, and a similar configuration may also be adopted for a teaching method for the atmospheric transfer device 42.

[0074] In the above embodiment, an example has been described in which the sensor 227 detects the step inner peripheral surface RS of the ring R as the edge of the circular measurement object. However, the edge of the ring R may be the outer edge of the outer annular surface Rout of the ring R, or the inner edge of the inner annular surface Rin of the ring R. Furthermore, the circular measurement object is not limited to the ring R, but may also be the substrate W or the substrate support part 12. For example, in the teaching method, by processing similar to that described above, the center position Wo of the substrate W can be calculated from measurement points on the outer edge of the substrate W, and the center position 12o can be calculated from measurement points on the outer edge of the substrate support surface 122a of the substrate support part 12.

[0075] Furthermore, when the fork 223 is moved forward and backward to detect a circular measurement target using the sensor 227, the control device 90 may detect four measurement points Rd when the fork 223 moves forward, and then detect four measurement points Rd when the fork 223 moves backward. By shifting the fork 223 laterally during this forward and backward movement, it is possible to change the positions of the measurement points Rd after the shift.

[0076] Furthermore, the calculation of the measured virtual circle and the extracted virtual circle is not limited to the least squares circle, and various calculation methods that can calculate a circle from multiple measurement points Rd may be used.Furthermore, the method of calculating the error between the measured virtual circle and each measurement point Rd is not limited to the least squares error, and various error calculation methods may be used.The method of calculating the error between the extracted virtual circle and each measurement point Rd of the outlier is not limited to the square error, and various error calculation methods may be used.

[0077] The technical ideas and effects of the present disclosure explained in the above embodiments will be described below.

[0078] A first aspect of the present disclosure is a teaching method for teaching the position of a circular measurement target in a semiconductor manufacturing apparatus (substrate processing system) including a transfer device (vacuum transfer device 22) having a fork 223 capable of transporting a substrate W and a control device 90 that controls the operation of the transfer device, the teaching method including: (A) a step of measuring the edge of the circular measurement target multiple times using a sensor 227 provided on the fork 223; (B) a step of creating a measurement virtual circle (measured least-squares circle I1 to I3) based on the multiple measurement points Rd measured by the sensor 227; (C) a step of calculating the error between the measurement virtual circle and the multiple measurement points Rd; (D) a step of creating an extracted virtual circle (extracted least-squares circle IE) by excluding outlier measurement points Rd from the multiple measurement points Rd based on the error; and (E) a step of comparing the extracted virtual circle with the excluded outlier measurement points Rd, and adding the outlier measurement points Rd that can be added to the extracted virtual circle to calculate the final center position of the circle.

[0079] As described above, the teaching method can accurately obtain the position of a circular measurement object by appropriately adding outlier measurement points Rd to create a circle, even when the number of data points is reduced by excluding outlier measurement points Rd. For example, it is possible to create an extracted least-squares circle IE with higher accuracy than the initial extracted least-squares circle IE using only measurement points Rd whose least-squares error is below a threshold. In particular, even when the number of data points is reduced, the teaching method can improve the accuracy of the extracted least-squares circle IE by using existing data without performing new measurements, thereby shortening the time required for the teaching method. Furthermore, for example, even when measurements are performed on a measurement object with deposits or wear, the teaching method can calculate the extracted least-squares circle IE from appropriate measurement points and accurately calculate the center position of the circle.

[0080] Furthermore, in step (B), a least-squares circle is created as a measured virtual circle (measured least-squares circle I1 to I3) based on the multiple measurement points Rd, and in step (D), a least-squares circle is created as an extracted virtual circle (extracted least-squares circle IE) based on the multiple measurement points Rd excluding outlier measurement points Rd. In this way, the teaching method can obtain measured virtual circles and extracted virtual circles with reduced influence of roundness and measurement variation based on the multiple measurement points.

[0081] In step (B), measurement virtual circles (measurement least-squares circles I1 to I3) are created based on the multiple measurement points Rd measured for each of the multiple measurements, and in step (C), the error between each measurement virtual circle and the multiple measurement points Rd is calculated. In this way, by obtaining a measurement virtual circle for each of the multiple measurements, the teaching method can recognize when an error occurred during the measurement and easily exclude measurement points Rd during measurements with large errors.

[0082] In step (C), the least square error of the measured virtual circle (measured least squares circle I1 to I3) is calculated as the error. This allows the teaching method to properly obtain the error of the measured virtual circle using the least squares error.

[0083] In step (D), a predetermined threshold is compared with the least squares error of the plurality of measured virtual circles (measured least squares circles I1 to I3), and an extracted virtual circle (extracted least squares circle IE) is created based on the plurality of measurement points Rd that form the measured virtual circle that is equal to or smaller than the threshold. This allows the teaching method to smoothly calculate the extracted virtual circle.

[0084] In step (E), the outlier measurement points Rd are added to the extracted virtual circle (extracted least-squares circle IE) in order of decreasing error among the removed outlier measurement points Rd, and it is determined whether the least-squares error of the extracted virtual circle after addition is equal to or less than a threshold. If the least-squares error of the extracted virtual circle is equal to or less than the threshold, the outlier measurement points Rd are added, and if the least-squares error of the extracted virtual circle exceeds the threshold, the outlier measurement points are not added. In this way, the teaching method adds the outlier measurement points Rd in order of decreasing error, making it possible to efficiently and accurately create an extracted virtual circle that includes the outlier measurement points Rd.

[0085] Furthermore, the transfer device (vacuum transfer device 22) is provided with a pair of support plate parts 225 with a width narrower than the diameter of the substrate W, and each of the pair of support plate parts 225 has a sensor 227, and in step (A), by moving the fork 223 forward and backward relative to the measurement object, four locations on the edge of the measurement object are measured as measurement points Rd by the sensor 227 of each of the pair of support plate parts 225. This makes it possible for the teaching method to easily obtain four measurement points Rd by measuring with one advancement of the fork 223.

[0086] Furthermore, in step (A), after each measurement, the fork 223 is shifted in a direction perpendicular to the direction in which the fork 223 advances and retreats, and after the shift, the fork 223 is advanced and retreated. This makes it possible for the teaching method to obtain measurement points Rd at different positions on the object to be measured, reducing local variations in the circularity of the object to further improve the accuracy of circle calculation.

[0087] Furthermore, the circular measurement target is a ring R that is arranged around the substrate W. This allows the teaching method to teach the position of the ring R easily and accurately.

[0088] The circular measurement target is the substrate W. Even in this case, the teaching method can teach the position of the substrate W easily and accurately.

[0089] Furthermore, the circular measurement target is the substrate support part 12 that supports the substrate W. Even in this case, the teaching method can teach the position of the substrate support part 12 easily and accurately.

[0090] A second aspect of the present disclosure is a semiconductor manufacturing apparatus (substrate processing system 1) including a transfer device (vacuum transfer device 22) having a fork 223 capable of transferring a substrate W and a control device 90 that controls the operation of the transfer device. When teaching the position of a circular measurement target, the control device 90 performs the following steps: (A) measuring the edge of the circular measurement target multiple times using a sensor 227 provided on the fork 223; (B) creating a measurement virtual circle (measured least-squares circles I1-I3) based on the multiple measurement points Rd measured by the sensor 227; (C) calculating the error between the measurement virtual circle and the multiple measurement points Rd; (D) creating an extracted virtual circle (extracted least-squares circle I1) by excluding outlier measurement points Rd from the multiple measurement points Rd based on the error; and (E) comparing the extracted virtual circle with the excluded outlier measurement points Rd, adding any outlier measurement points Rd that can be added to the extracted virtual circle, and calculating the final center position of the circle. Even in this case, the semiconductor manufacturing apparatus can accurately obtain the position of the circular measurement target.

[0091] The teaching method and semiconductor manufacturing apparatus according to the presently disclosed embodiments are illustrative in all respects and not restrictive. The embodiments may be modified and improved in various ways without departing from the spirit and scope of the appended claims. The features described in the above embodiments may be configured in other ways as long as they are not inconsistent, and may be combined as long as they are not inconsistent. [Explanation of symbols]

[0092] 22 Vacuum transport device 90 Control device 223 Fork 227 Sensors IE Extracting Least Squares Circle I1~I3 Measured least squares circle Rd measurement point W substrate

Claims

1. a transport device having a fork capable of transporting a substrate; a control device that controls the operation of the transport device, a teaching method for teaching a position of a circular measurement target in a semiconductor manufacturing apparatus, the teaching method comprising: (A) measuring the edge of the circular measurement target multiple times using a sensor provided on the fork; (B) creating a measurement virtual circle based on a plurality of measurement points measured by the sensor; (C) calculating an error between the measurement virtual circle and the plurality of measurement points; (D) creating an extracted virtual circle by excluding outlier measurement points from the plurality of measurement points based on the error; (E) comparing the extracted virtual circle with the excluded measurement points of the outliers, and adding the measurement points of the outliers that can be added to the extracted virtual circle to calculate a final center position of the circle. Teaching methods.

2. In the step (B), a least-squares circle is created as the measurement virtual circle based on the plurality of measurement points; In the step (D), a least-squares circle is created as the extracted virtual circle based on the plurality of measurement points excluding the outlier measurement points. The teaching method according to claim 1 .

3. In the step (B), the measurement virtual circle is created based on the plurality of measurement points measured for each of the plurality of measurements; In the step (C), an error between each of the measurement virtual circles and the plurality of measurement points is calculated. The teaching method according to claim 2 .

4. In the step (C), a least square error of the measurement virtual circle is calculated as the error. The teaching method according to any one of claims 1 to 3.

5. In the step (D), a minimum square error of the plurality of measurement virtual circles is compared with a threshold value stored in advance, and the extracted virtual circle is created based on the plurality of measurement points that form the measurement virtual circle that is equal to or smaller than the threshold value. The teaching method according to claim 4.

6. In the step (E), the measurement points of the excluded outliers are added to the extracted virtual circle in ascending order of error, and it is determined whether or not the least square error of the extracted virtual circle after addition is equal to or less than the threshold value. If the least square error of the extracted virtual circle is equal to or less than the threshold value, the measurement points of the outliers are added, and if the least square error of the extracted virtual circle exceeds the threshold value, the measurement points of the outliers are not added. The teaching method according to claim 5 .

7. the transport device includes a pair of support plate portions having a width narrower than a diameter of the substrate, and the sensor is provided on each of the pair of support plate portions; In the step (A), the fork is moved forward and backward relative to the measurement object, and four points on the edge of the measurement object are measured as the measurement points by the sensors of each of the pair of support plate portions. The teaching method according to any one of claims 1 to 3.

8. In the step (A), the fork is shifted in a direction perpendicular to the direction of advancement and retreat of the fork every time the measurement is performed a plurality of times, and the fork is advanced and retreated after the shift. The teaching method according to claim 7.

9. the circular measurement target is a ring disposed around the substrate; The teaching method according to any one of claims 1 to 3.

10. the circular measurement target is the substrate; The teaching method according to any one of claims 1 to 3.

11. the circular measurement target is a substrate support portion that supports the substrate; The teaching method according to any one of claims 1 to 3.

12. a transport device having a fork capable of transporting a substrate; a control device for controlling the operation of the transport device, When teaching the position of a circular measurement target, the control device (A) measuring the edge of the circular measurement target multiple times using a sensor provided on the fork; (B) creating a measurement virtual circle based on a plurality of measurement points measured by the sensor; (C) calculating an error between the measurement virtual circle and the plurality of measurement points; (D) creating an extracted virtual circle by excluding outlier measurement points from the plurality of measurement points based on the error; (E) comparing the extracted virtual circle with the measurement points of the excluded outliers, and adding the measurement points of the outliers that can be added to the extracted virtual circle to calculate the final center position of the circle; Semiconductor manufacturing equipment.

Images