Method for measuring defect height of epitaxial wafer and defect height measuring apparatus

Abstract

To provide a defect height measuring method of an epitaxial wafer capable of efficiently measuring the height of a defect.SOLUTION: A method for measuring a defect height of an epitaxial wafer is a method for measuring a defect height of an epitaxial wafer performed by controlling a microscope by a control device, and the control device performs a review range setting step of setting a review range that includes a defect whose area in a range setting image of a measurement target range of the epitaxial wafer imaged by the microscope is equal to or larger than a range threshold and is smaller than the measurement target range, and a height measuring step of measuring a height of the defect based on a measurement image of the review range imaged by the microscope at a magnification higher than that of the range setting image.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to a defect height measuring method and a defect height measuring device for an epitaxial wafer. [Background technology]

[0002] Conventionally, configurations for measuring defects on the surface of a wafer are known (see, for example, Patent Documents 1 and 2). The defects to be measured in the configuration of Patent Document 1 are defects (hereinafter sometimes referred to as "pin mark defects") caused by the shape of the lift pins of the vapor phase growth apparatus or the contact state between the lift pins and the wafer during epitaxial growth. Pin mark defects are point-like defects consisting of an aggregate of minute scratches such as wear marks or deposits that are difficult to remove, such as Si coating components on the lift pins of the vapor phase growth apparatus. Multiple pin mark defects occur densely in the contact area between the lift pin and the epitaxial wafer. The measurement method described in Patent Document 1 calculates the area of ​​multiple point-like defects present on the epitaxial wafer based on the imaging results of the imaging unit. Then, pin mark defects are detected based on the calculation result of the area of ​​the point-like defects. In the measurement method described in Patent Document 2, a roughness measuring device such as a white light interference microscope is used to measure the roughness of the wafer surface. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2018-146482 [Patent Document 2] Japanese Patent Publication No. 2023-41001 Summary of the Invention [Problem to be solved by the invention]

[0004] As described in Patent Document 1, pin mark defects may cause defocusing in the photolithography process included in the device formation process. In order to suppress the occurrence of such defocusing, it is preferable to measure the height of the pin mark defects and eliminate defects that are tall enough to cause defocusing. However, the measurement method of Patent Document 1 can only evaluate the area of ​​pin mark defects, and therefore cannot eliminate defects that are tall enough to cause defocus. Furthermore, in the measurement method of Patent Document 2, the measurement range is narrow, so unless the position of the defect is specified in advance, there is a risk that the measurement cannot be performed efficiently.

[0005] An object of the present invention is to provide a defect height measuring method and a defect height measuring device for an epitaxial wafer that can efficiently measure the height of a defect. [Means for solving the problem]

[0006] The epitaxial wafer defect height measurement method of the present invention is a method for measuring the height of a defect in an epitaxial wafer performed by a control device controlling a microscope, wherein the control device performs a review range setting process for setting a review range that includes defects whose area in a range setting image of a measurement range of the epitaxial wafer captured by the microscope is equal to or greater than a range threshold and is smaller than the measurement range, and a height measurement process for measuring the height of the defect based on a measurement image of the review range captured by the microscope at a magnification higher than that of the range setting image.

[0007] In the epitaxial wafer defect height measurement method of the present invention, it is preferable that the review range setting process includes a low-magnification review process in which a first review range is set so that an area in a first range setting image of the measurement target range, taken at a first magnification, includes defects whose area is equal to or greater than the first range threshold and is smaller than the measurement target range, and a high-magnification review process in which a second review range is set so that an area in a second range setting image of the first review range, taken at a second magnification higher than the first magnification, includes defects whose area is equal to or greater than the second range threshold and is smaller than the first review range, and the height measurement process measures the height of the defects based on the measurement image of the second review range taken at a third magnification higher than the second magnification.

[0008] In the epitaxial wafer defect height measurement method of the present invention, the low magnification review process preferably includes a first setting process of setting the first review range to include the defect with the largest area among the defects whose area in the first range setting image is equal to or greater than the first range threshold, and a second setting process of setting the first review range to include the defect with the largest area among the defects whose area in the first range setting image is equal to or greater than the first range threshold and which are not included in the first review range that has already been set, and the second setting process is preferably repeated until a first number of first review ranges have been set.

[0009] In the epitaxial wafer defect height measurement method of the present invention, the high-magnification review process preferably includes a third setting process of setting the second review range to include the defect with the largest area among the defects whose area in the second range setting image is equal to or greater than the second range threshold, and a fourth setting process of setting the second review range to include the defect with the largest area among the defects whose area in the second range setting image is equal to or greater than the second range threshold and which are not included in the second review range that has already been set, and the fourth setting process is preferably repeated until the second number of ranges of second review ranges have been set.

[0010] In the epitaxial wafer defect height measurement method of the present invention, it is preferable that the high-magnification review process further includes an exclusion process of excluding the first review range in which the number of defects equal to or greater than the second range threshold included in the second range setting image is less than an exclusion reference value from the targets for setting the second review range in the third setting process and the fourth setting process.

[0011] In the epitaxial wafer defect height measurement method of the present invention, it is preferable that the height measurement process includes a measurement target identification process for identifying a defect in the measurement image whose area is equal to or greater than a target threshold as a measurement target defect, and a calculation process for calculating the height of the measurement target defect based on a plurality of measurement images taken after the measurement target identification process from different height positions relative to the epitaxial wafer.

[0012] In the epitaxial wafer defect height measurement method of the present invention, it is preferable that the first range setting image and the second range setting image are captured by an optical microscope constituting the microscope, and the measurement image is captured by a white light interference microscope or a laser microscope constituting the microscope.

[0013] The epitaxial wafer defect height measurement device of the present invention comprises a microscope and a control device that controls the microscope to measure the height of a defect in the epitaxial wafer, and the control device comprises: a review range setting unit that sets a review range that includes defects whose area in a range setting image of the measurement target range of the epitaxial wafer captured by the microscope is equal to or greater than a range threshold and is smaller than the measurement target range; and a height measurement unit that measures the height of the defect based on a measurement image of the review range captured by the microscope at a higher magnification than the range setting image.

[0014] In the epitaxial wafer defect height measurement device of the present invention, it is preferable that the review range setting unit performs a low-magnification review process in which a first review range is set so that the area in a first range setting image of the measurement target range, taken at a first magnification, includes the defect whose area is equal to or greater than the first range threshold and is smaller than the measurement target range, and a high-magnification review process in which a second review range is set so that the area in a second range setting image of the first review range, taken at a second magnification higher than the first magnification, includes the defect whose area is equal to or greater than the second range threshold and is smaller than the first review range, and the height measurement unit measures the height of the defect based on the measurement image of the second review range taken at a third magnification higher than the second magnification.

[0015] In the epitaxial wafer defect height measuring device of the present invention, it is preferable that the microscope is composed of a height measuring microscope consisting of a white light interference microscope or a laser microscope, and an optical microscope, and that the first range setting image and the second range setting image are captured by the optical microscope, and the measurement image is captured by the height measuring microscope.

[0016] In the epitaxial wafer defect height measuring device of the present invention, it is preferable that a moving unit is provided for moving the microscope, and the control device controls the microscope and the moving unit to measure the height of the defect in the epitaxial wafer. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a schematic diagram showing the configuration of an epitaxial wafer defect height measuring device according to an embodiment. [Figure 2] 1 is a block diagram showing a configuration of a defect height measuring device according to an embodiment. [Figure 3] 1A and 1B are explanatory diagrams of a defect height measuring method according to an embodiment. [Figure 4] 1 is a flowchart of a defect height measuring method according to an embodiment. [Figure 5] 10 is a flowchart of a low magnification review process according to the embodiment. [Figure 6] 10 is a flowchart of a low magnification review process according to the embodiment. [Figure 7] 10 is a flowchart of a high-magnification review process according to the embodiment. [Figure 8] 10 is a flowchart of a high-magnification review process according to the embodiment. [Figure 9] 10 is a flowchart of a height measurement process according to the embodiment. [Figure 10] 10 is a flowchart of a defect height measuring method according to a modified example of the present invention. [Figure 11] 10 is a graph showing the correlation between the maximum values ​​of height measurement results obtained by automatic measurement and manual measurement for each epitaxial wafer according to Experimental Example 1 of the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0018] [Embodiment] An embodiment of the present invention will be described below. <Configuration of epitaxial wafer defect height measurement device> First, the configuration of the epitaxial wafer defect height measuring device according to this embodiment will be described. Fig. 1 is a schematic diagram showing the configuration of the epitaxial wafer defect height measuring device. Fig. 2 is a block diagram showing the configuration of the defect height measuring device. Fig. 3 is an explanatory diagram of a defect height measuring method.

[0019] 1 measures the height of defects on the back surface W1, which is the other main surface of the epitaxial wafer W. The defects to be measured in this embodiment are pin mark defects that occur due to the shape of the lift pins of the vapor phase growth apparatus and the contact state between the lift pins and the epitaxial wafer W during epitaxial growth, as described above. The diameter of the epitaxial wafer W is not particularly limited. The defect height measuring device 1 includes a mounting table 2, an aligner 3, a measurement stage 4, an observation unit 5, a moving unit 6 (see FIG. 2), a transfer robot 7, and a control device 8.

[0020] The mounting table 2, the aligner 3, the measurement stage 4, the observation unit 5, and the transfer robot 7 are arranged inside a housing 10. A load port 11 is detachably attached to the front of the housing 10. The load port 11 stores a plurality of epitaxial wafers W with their back surfaces W1 facing downward. An opening is provided in the front of the housing 10 so that the epitaxial wafers W can be transported between the inside of the housing 10 and the inside of the load port 11.

[0021] The table 2 has four placement portions 21 arranged to form an imaginary circle. An epitaxial wafer W is placed on the upper surface of each mounting portion 21 with the front surface W2 facing upward. The upper surface of each mounting portion 21 is inclined so that the height position decreases as it approaches the center of the epitaxial wafer W. Therefore, the epitaxial wafer W is placed on the mounting table 2 with only the outer edge of the back surface W1 in contact with the mounting portion 21. By placing the epitaxial wafer W on the mounting table 2 as described above, it is possible to prevent dust from adhering to the back surface W1 of the epitaxial wafer W and the occurrence of scratches.

[0022] The aligner 3 includes four mounting portions 31 and four grips 32 arranged in an imaginary circle. An epitaxial wafer W is placed on the upper surface of each mounting portion 31 with its back surface W1 facing upward. The upper surface of each mounting portion 31 is inclined in the same manner as the mounting portion 21 of the stage 2. Therefore, the epitaxial wafer W is placed on the mounting portion 31 with only the outer edge of its front surface W2, which serves as one of its main surfaces, in contact with the mounting portion 31. The four grips 32 are disposed between adjacent mounting portions 31. Based on the control of the control device 8, each grip 32 holds only the outer edge of the epitaxial wafer W and adjusts the position of the epitaxial wafer W on the mounting portion 31 so that the notch WN of the epitaxial wafer W is located at a predetermined position. By placing the epitaxial wafer W on the mounting portion 31 and holding the epitaxial wafer W by the grip 32 as described above, it is possible to prevent dust from adhering to the back surface W1 or the front surface W2 of the epitaxial wafer W and prevent scratches from occurring.

[0023] The measurement stage 4 holds the epitaxial wafer W so that the back surface W1 faces upward. The measurement stage 4 includes a wafer holder 41, and a lift pin driver and a clamp driver (not shown) that are driven under the control of the control device 8. The wafer holder 41 includes two supports 42, four stage lift pins 43, and four clamps 44.

[0024] The upper surface of each support 42 is inclined so that the height position becomes lower as it approaches the center of the epitaxial wafer W. Therefore, the epitaxial wafer W is placed on the support 42 with only the outer edge of the front surface W2 in contact with the upper surface.

[0025] Each stage lift pin 43 is raised and lowered by being driven by a lift pin drive unit to receive an epitaxial wafer W before measurement from the transfer robot 7 and to deliver an epitaxial wafer W after measurement to the transfer robot 7. The mounting surface for the epitaxial wafer W on the upper surface of each stage lift pin 43 is inclined so that the height position becomes lower as it approaches the center of the epitaxial wafer W. Therefore, the epitaxial wafer W is supported by the stage lift pins 43 with only the outer edge of the front surface W2 in contact with the mounting surface.

[0026] Each clamp 44 has a clamp groove (not shown) extending along the outer periphery. Of the clamps 44, the two clamps 44 on the right side in FIG. 1 constitute a first set of clamps 44A, and the two clamps 44 on the left side constitute a second set of clamps 44B. The first set of clamps 44A and the second set of clamps 44B are driven by a clamp drive unit to move toward each other in the left-right direction, and the outer edges of the back surface W1 and front surface W2 of the epitaxial wafer W come into contact with each clamp groove, thereby clamping the epitaxial wafer W supported by the support 42.

[0027] The observation unit 5 includes an optical microscope 51 and a white light interference microscope 52 as a microscope for measuring height. The optical microscope 51 captures an image of the measurement target area of ​​the epitaxial wafer W and outputs the image to the control device 8. The measurement target area is the contact area with the lift pins of the vapor phase growth apparatus, and if the vapor phase growth apparatus has three lift pins, there are three measurement target areas. The white light interference microscope 52 captures an image of each measurement target range at a higher magnification than the optical microscope 51, and outputs to the control device 8 an image used to measure the height of a defect. Instead of the white light interference microscope 52, a laser microscope or the like may be used as a height measuring microscope to measure the height of the defect.

[0028] The moving unit 6 moves the optical microscope 51 and the white light interferometer 52 up and down independently under the control of the control device 8. The moving unit 6 moves the measurement stage 4 up and down and back and forth under the control of the control device 8. With this configuration, the measurement stage 4 moves from the standby position shown in FIG. 1 to below the optical microscope 51 or the white light interferometer 52, and the optical microscope 51 or the white light interferometer 52 moves up and down, so that a desired position on the epitaxial wafer W can be observed.

[0029] The transfer robot 7 includes an upper hand 71 that is driven under the control of the control device 8, and a lower hand (not shown) that is positioned below the upper hand 71. Chucks 72 are provided at the tip and base of the upper surface of the upper hand 71. The upper hand 71 is configured to be extendable and rotatable, and the multiple chucks 72 enable the semiconductor wafer W to be inverted while holding only the outer edge of the semiconductor wafer W. Chucks (not shown) are provided at the tip and base of the upper surface of the lower hand, similar to the upper hand 71. The lower hand can hold only the outer edge of the semiconductor wafer W by means of a plurality of chucks. Since the upper hand 71 and the lower hand have the above-described configuration, the back surface W1 and the front surface W2 of the epitaxial wafer W are prevented from being contaminated with dust or from being scratched.

[0030] 2, the control device 8 is configured to be able to transmit and receive various types of information to and from the aligner 3, the measurement stage 4, the observation unit 5, the moving unit 6, and the transport robot 7. The control device 8 includes an input unit 81, a display unit 82, an output unit 83, a memory unit 84, and a control unit 85.

[0031] The input unit 81 is configured by, for example, a touch panel or physical buttons, and outputs a signal corresponding to an input operation to the control unit 85. The display unit 82 displays various information under the control of the control unit 85. The output unit 83 is connected to an external device such as an operation system (not shown) so as to be able to output various information.

[0032] The storage unit 84 stores various information related to the height measurement of the epitaxial wafer W so that the information can be read by the control unit 85.

[0033] The control unit 85 includes a CPU, and realizes various functions by the CPU executing programs stored in the storage unit 84. The control unit 85 includes a review range setting unit 851, a height measurement unit 852, and an output control unit 853.

[0034] 3, the review range setting unit 851 sets a first review range A1 in which the area in the first range setting image P1 of the measurement target range A0 of the epitaxial wafer W captured at the first magnification by the optical microscope 51 includes defects D equal to or greater than the first range threshold and is smaller than the measurement target range A0. The measurement target range A0 is set to an area including a contact portion with the lift pin of the vapor phase growth apparatus. In this embodiment, three measurement target ranges A0 are set. In FIG. 3, some of the defects D are shown as being round, but they are not limited to being round and may have a variety of shapes.

[0035] The review range setting unit 851 sets a second review range A2 whose area in the second range setting image P2 of the first review range A1 captured by the optical microscope 51 at a second magnification higher than the first magnification includes defects D equal to or greater than the second range threshold and is smaller than the first review range A1.

[0036] The height measurement unit 852 measures the height of the defect D included in the second review range A2 based on the measurement image P3 of the second review range A2 captured by the white light interference microscope 52 at a third magnification higher than the second magnification.

[0037] The output control unit 853 controls the output unit 83 to output the height measurement result from the height measurement unit 852 to an external device.

[0038] <Method for measuring defect height in epitaxial wafers> Next, a defect height measurement method for an epitaxial wafer W using the defect height measurement device 1 will be described. FIG. 4 is a flowchart of the defect height measurement method. FIGS. 5 and 6 are flowcharts of the low-magnification review process. FIGS. 7 and 8 are flowcharts of the high-magnification review process. FIG. 9 is a flowchart of the height measurement process.

[0039] The lower hand of the transfer robot 7 enters below the semiconductor wafer W in the load port 11, holds the outer edge with a plurality of chucks, and places the semiconductor wafer W on the table 2 with the front surface W2 facing upward. The upper hand 71 of the transport robot 7 enters below the semiconductor wafer W placed on the table 2 and holds the semiconductor wafer W with the front surface W2 facing upward with a plurality of chucks 72. The transport robot 7 extends the upper hand 71 to invert the semiconductor wafer W, transports it, and places it on the placement section 31 of the aligner 3 with the back surface W1 facing upward. The lower hand then moves below the semiconductor wafer W whose position has been adjusted by the aligner 3, and holds it with multiple chucks. While maintaining the state after the position adjustment, the lower hand moves the epitaxial wafer W with the back surface W1 facing upward, above the measurement stage 4, which is in a clamp standby state.

[0040] After each stage lift pin 43 has risen to a position higher than the support 42, the transfer robot 7 descends, releases the epitaxial wafer W, and places the epitaxial wafer W on each stage lift pin 43. After placing the epitaxial wafer W, the transfer robot 7 moves to a position away from above the measurement stage 4. Each stage lift pin 43 is lowered to a position lower than each support 42 , and the epitaxial wafer W is placed on each support 42 .

[0041] The first set of clamps 44A and the second set of clamps 44B move toward each other. As the clamps 44 move, the epitaxial wafer W moves away from the supports 42 and is clamped. The control device 8 controls the observation unit 5 to measure the height of the defect D present in the epitaxial wafer W. The specific measurement method will be described later.

[0042] When the measurement of the height of the defect D is completed, the first set of clamps 44A and the second set of clamps 44B move in a direction away from the epitaxial wafer W. As each clamp 44 moves, the epitaxial wafer W is released from the clamps 44 and is supported by each support 42.

[0043] Each stage lift pin 43 rises to a position higher than the epitaxial wafer W and receives the epitaxial wafer W from each support 42 . The upper hand 71 of the transfer robot 7 enters below the semiconductor wafer W supported by each stage lift pin 43 and holds the semiconductor wafer W with a plurality of chucks 72. The transfer robot 7 extends and inverts the upper hand 71, and places the semiconductor wafer W on the stage 2 with the front surface W2 facing upward. The lower hand enters below the semiconductor wafer W placed on the stage 2, holds the semiconductor wafer W with the front surface W2 facing upward with a plurality of chucks, and stores the semiconductor wafer W in the load port 11. After receiving the epitaxial wafer W, each stage lift pin 43 is lowered to its pre-lift position.

[0044] Next, the defect height measuring method will be described in detail. As shown in FIG. 4, the review range setting unit 851 of the control device 8 performs a low-magnification review process (step S1). In the low-magnification review process, the review range setting unit 851 specifies one measurement range A0 on the back surface W1 of the epitaxial wafer W, as shown in FIG. 5 (step S11). In step S11, the review range setting unit 851 sets the measurement range A0 based on coordinates indicating the center of the measurement range A0 that has been set in advance by an operator operating the input unit 81, for example. The measurement range A0 does not have to be square, and may be other shapes such as a circle, etc. An example of the shape of the measurement range A0 is a square with one side measuring 20 mm.

[0045] The review range setting unit 851 controls the observation unit 5 to obtain a divided image P10 shown in FIG. 3 (step S12). In step S12, the review range setting unit 851 sets the magnification of the optical microscope 51 to a first magnification. The magnification of the optical microscope 51 may be set by the operator by manipulating the objective lens. The first magnification in this embodiment is set to 2.5 times. The review range setting unit 851 controls the moving unit 6 to move the measurement stage 4 horizontally while the optical microscope 51 is positioned above the measurement target range A0, and acquires divided images P10 of the multiple divided ranges that make up the measurement target range A0.

[0046] The review range setting unit 851 generates a first range setting image P1 by combining a plurality of divided images P10 so as to represent a bird's-eye view of the entire measurement target range A0 (step S13).

[0047] The review range setting unit 851 generates a first difference image (step S14). In step S14, the review range setting unit 851 generates a first average image by averaging the first range setting image P1 using, for example, an averaging filter, and generates a first difference image that represents the difference in brightness between the first average image and the first range setting image P1.

[0048] The review range setting unit 851 identifies the first detected defect D1 (step S15). In step S15, the review range setting unit 851 generates a binary image of the first difference image, and identifies at least one area in the binary image whose brightness exceeds the first brightness threshold and whose area is equal to or greater than the first range threshold as the first detected defect D1 among the areas corresponding to defect D.

[0049] As shown in FIG. 6, the review range setting unit 851 uses well-known image processing to select the first detected defect D1 with the largest area from among the first detected defects D1 identified in step S15 (step S16: first setting step).

[0050] The review range setting unit 851 sets a first review range A1 that includes the first detected defect D1 selected in step S16 (step S17: first setting step). In step S17, the review area setting unit 851 sets a square area of ​​a predetermined size that includes the selected first detected defect D1 as a first review area A1. Note that first review area A1 may have a shape other than a square, such as a circle, etc. Also, first review area A1 may be set so that the selected first detected defect D1 is located at the center.

[0051] The review range setting unit 851 determines whether a first review range A1 of the first range number has been set (step S18). The first range number is not particularly limited, but can be set based on the viewpoint of increasing the measurement accuracy by increasing the number of measurement target candidates, and the viewpoint that if there are too many measurement target candidates, the measurement time will be long. In this embodiment, the first range number is set to "12."

[0052] In step S18, if the review range setting unit 851 determines that the first number of first review ranges A1 have not been set (step S18: NO), it uses well-known image processing to select the first detected defect D1 with the largest area from among the first detected defects D1 that are not included in the already set first review ranges A1 (step S19: second setting process). Similar to the processing of step S17, the review range setting unit 851 sets a first review range A1 that includes the first detected defect D1 selected in step S19 (step S20: second setting step), and performs the processing of step S18. In step S18, if the review range setting unit 851 determines that the first number of first review ranges A1 have been set (step S18: YES), the low magnification review process ends. By the above-described low-magnification review process, twelve first review areas A1 are set, as indicated by the two-dot chain line squares in the upper left diagram of FIG.

[0053] In the processing of steps S18 to S20 described above, for example, if the first review range A1 set includes the first and second largest detected defects D1, the second first review range A1 is set based on the third largest detected defect D1 rather than the second largest detected defect D1. This allows the first number of first review ranges A1 to include more tall first detected defects D1 than in a configuration in which the second largest detected defect D1 included in the first first review range A1 is used as the reference when setting the second first review range A1.

[0054] 4, after the low-magnification review process is completed, the review range setting unit 851 performs a high-magnification review process (step S2). The low-magnification review process and the high-magnification review process constitute a review range setting process.

[0055] In the high-magnification review process, the review range setting unit 851 controls the observation unit 5 to acquire a second range setting image P2, as shown in FIG. 7 (step S31). In step S31, the review range setting unit 851 sets the magnification of the optical microscope 51 to a second magnification that is higher than the first magnification. The magnification of the optical microscope 51 may be set by the operator by manipulating the objective lens. The second magnification in this embodiment is set to 20 times. The review range setting unit 851 controls the moving unit 6 to move the measurement stage 4 horizontally with the optical microscope 51 positioned above each first review range A1, while acquiring multiple second range setting images P2 representing each first review range A1. In this embodiment, the number of second range setting images P2 is "12," the same as the number of first review ranges A1.

[0056] The review range setting unit 851 generates a second composite image by combining the plurality of second range setting images P2 so as to represent a bird's-eye view of all of the first review ranges A1 (step S32).

[0057] The review range setting unit 851 generates a second difference image (step S33). In step S33, the review range setting unit 851 calculates the average value of the luminance of the second composite image, generates a second average image in which the overall luminance value is set to the average value, and generates a second difference image that represents the difference in luminance between the second average image and the second composite image. Note that the second average image may be generated using an averaging filter, similar to the process of generating the first average image in step S14. In the process of generating the first average image in step S14, the first average image may be generated by performing a process similar to the process of generating the second average image in step S33.

[0058] The review range setting unit 851 identifies the second detected defect D2 (step S34). In step S34, the review range setting unit 851 generates a binary image of the second difference image, and identifies as the second detected defect D2 at least one area in which the brightness in the binary image exceeds the second brightness threshold and which corresponds to the first detected defect D1 and whose area is equal to or greater than the second range threshold.

[0059] The review range setting unit 851 determines the first review range A1 to be excluded from the setting targets of the second review range A2 (step S35: exclusion step). In step S35, the review range setting unit 851 calculates the number of second detected defects D2 for each first review range A1, and excludes any first review range A1 in which the number of second detected defects D2 is less than an exclusion threshold value from the set of second review ranges A2. The exclusion threshold value can be, for example, a value between 1 and 200.

[0060] Here, the defects D to be measured are pin mark defects that occur in close proximity, as described above. If the number of second detected defects D2 included in the first review range A1 is less than the exclusion reference value, the second detected defects D2 are unlikely to be pin mark defects and are likely to be easily removed deposits such as dust. The above-described exclusion process in step S35 allows defects D that are unlikely to be pin mark defects to be excluded from the measurement target, improving measurement efficiency.

[0061] As shown in FIG. 8, the review range setting unit 851 uses well-known image processing to select the second detected defect D2 with the largest area from among the second detected defects D2 identified in step S34 (step S36).

[0062] The review range setting unit 851 sets a second review range A2 that includes the second detected defect D2 selected in step S36 (step S37: third setting step). In step S37, the review area setting unit 851 sets a square area of ​​a predetermined size that includes the selected second detected defect D2 as a second review area A2. The second review range A2 may have a shape other than a square, such as a circle, etc. The second review range A2 may also be set so that the selected second detected defect D2 is located at the center.

[0063] The review range setting unit 851 determines whether a second review range A2 of a second range number has been set (step S38). The second range number can be set based on the same considerations as the first range number. In this embodiment, the second range number is set to "12." The second number of ranges may be the same as the first number of ranges, or may be different.

[0064] In step S38, if the review range setting unit 851 determines that the second number of second review ranges A2 have not been set (step S38: NO), it uses well-known image processing to select the second detected defect D2 with the largest area from among the second detected defects D2 that are not included in the already set second review ranges A2 (step S39). Similar to the processing in step S37, the review range setting unit 851 sets a second review range A2 that includes the second detected defect D2 selected in step S39 (step S40: fourth setting step), and performs the processing in step S38. In step S38, if the review range setting unit 851 determines that the second number of second review ranges A2 have been set (step S38: YES), the high magnification review process ends. By the above high-magnification review process, 12 second review areas A2 are set, as shown by the dashed-dotted squares in the upper left diagram of FIG.

[0065] In the processing of steps S38 to S40 described above, for example, if the first set second review range A2 includes the second detected defects D2 with the first and second largest areas, the second review range A2 is set based on the second detected defect D2 with the third largest area, rather than the second detected defect D2 with the second largest area. This allows the second number of second review ranges A2 to include more tall second detected defects D2 than in a configuration in which the second detected defect D2 with the second largest area included in the first second review range A2 is used as the base when setting the second second review range A2.

[0066] As shown in FIG. 4, when the high-magnification review process is completed, the height measurement unit 852 performs a height measurement process (step S3).

[0067] In the height measurement step, the height measurement unit 852 selects a second review range A2 that includes a second detected defect D2 with the largest area, as shown in FIG. 9 (step S51).

[0068] The height measurement unit 852 controls the observation unit 5 to acquire a measurement image P3 (step S52). In step S52, the height measurement unit 852 sets the magnification of the white light interference microscope 52 to a third magnification that is higher than the second magnification. The magnification of the white light interference microscope 52 may be set by the operator by manipulating the objective lens. The third magnification in this embodiment is set to 100 times. The height measurement unit 852 controls the movement unit 6 to acquire measurement images P3 representing the second review range A2 while moving the measurement stage 4 horizontally with the white light interferometer 52 positioned above the second review range A2 selected in step S51. The number of measurement images P3 in this embodiment is "12," the same as the number of second review ranges A2.

[0069] The height measurement unit 852 generates a third difference image (step S53). In step S53, the height measurement unit 852 generates a third average image by performing, for example, the same process as the process for generating the first average image in step S14 or the process for generating the second average image in step S33. The height measurement unit 852 generates a third difference image that represents the difference in brightness between the third average image and measurement image P3.

[0070] The height measurement unit 852 identifies the second detected defect D2 of the measurement target as the measurement target defect Da (step S54: measurement target identification step). In step S54, the height measurement unit 852 generates a binary image of the third difference image, and identifies the area in the binary image corresponding to the second detected defect D2, whose brightness exceeds the third brightness threshold, and whose area is equal to or greater than the third area threshold, as the measurement target defect Da. In this way, by specifying only the region whose area is equal to or greater than the third area threshold, that is, the second detected defect D2 having a relatively large area, as the measurement target defect Da, the measurement efficiency is further improved.

[0071] The height measurement unit 852 controls the observation unit 5 to calculate the height of the measurement target defect Da (step S55: calculation step). In step S55, the height measurement unit 852 controls the movement unit 6 to move the optical system of the white light interference microscope 52 up and down to calculate the height of the measurement target defect Da, and stores the calculation result in the storage unit 84.

[0072] The height measurement unit 852 determines whether or not all of the second review ranges A2 have been selected in step S51 (step S56). In step S56, if the height measurement unit 852 determines that all second review ranges A2 have not been selected (step S56: NO), it selects the second review range A2 that includes the second detected defect D2 with the largest area from among the unselected second review ranges A2 using well-known image processing (step S57), and performs the processing of step S52.

[0073] In step S56, if the height measurement unit 852 determines that all second review ranges A2 have been selected (step S56: YES), it determines the largest calculated value among the values ​​calculated in step S55 as the height measurement value in the measurement target range A0 (step S58) and stores the height measurement value in the memory unit 84.

[0074] As shown in FIG. 4, when the height measurement process is completed, the review range setting unit 851 determines whether height measurement has been completed for all measurement ranges A0, that is, whether the low-magnification review process, high-magnification review process, and height measurement process of steps S1 to S3 have been performed (step S4). In step S4, if the review range setting unit 851 determines that height measurement has not been completed for all of the measurement target ranges A0 (step S4: NO), it performs the process of step S1. In step S4, if the review range setting unit 851 determines that height measurement has been completed for all measurement target ranges A0 (step S4: YES), the output control unit 853 performs a height measurement result output step (step S5). In the height measurement result output step, the output control unit 853 controls the output unit 83 to output the height measurement value to an external device. With the above, the height measurement process by the control device 8 is completed.

[0075] As described above, the control unit 85 of the control device 8 sets a first review range A1 that includes a first detected defect D1 whose area in the first range setting image P1 of the measurement target range A0 is equal to or greater than the first range threshold, and is smaller than the measurement target range A0. The control unit 85 sets a second review range A2 that includes a second detected defect D2 whose area in the second range setting image P2 of the first review range A1 is equal to or greater than the second range threshold, and is smaller than the first review range A1. The control unit 85 measures the height of the measurement target defect Da based on the measurement image P3 of the second review range A2. In this way, based on the areas of the first and second detected defects D1, D2 in the first and second range setting images P1, P2, the observation range in the observation unit 5 is gradually narrowed from the measurement target range A0 to the second review range A2, thereby narrowing down to some extent the measurement target defects Da to be subjected to height measurement from among the defects D present in the epitaxial wafer W. Therefore, the height of the defect D can be measured efficiently. In addition, the time required for inspection can be shortened, thereby improving the productivity of epitaxial wafers W. Furthermore, the ability to eliminate high defects can contribute to improving yields in customers' semiconductor device manufacturing processes. Furthermore, since an optical microscope 51 is used in the low-magnification review process and high-magnification review process of step S1, a wide area can be observed at a time at a relatively low magnification, for example, 2.5x or 20x, thereby improving the processing efficiency of the low-magnification review process and high-magnification review process.

[0076] [Variations] Although an embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and various improvements and design changes that do not deviate from the gist of the present invention are also included in the present invention.

[0077] As shown in FIG. 10, a pin scratch height determination step (step S61) and a pin scratch defect determination step (step S62) may be performed between the processes of steps S4 and S5 in FIG. In the pinch scratch height determination process, if the review range setting unit 851 determines that height measurements for all measurement ranges A0 have been completed (step S4: YES), the height measurement unit 852 determines the largest height measurement value among the height measurement values ​​determined in step S58 in all measurement ranges A0 of one epitaxial wafer W as the pinch scratch height on that epitaxial wafer W. In the pinch defect determination step, if the pinch defect height determined in the pinch defect height determination step exceeds a preset threshold, the height measurement unit 852 determines that the epitaxial wafer W to be measured is pinch defect. An example of the threshold in the pinch defect determination step is a value of 2 μm or more. In the measurement result output process of step S5, the output control unit 853 controls the output unit 83 to output the height measurement value, pinch defect height, and pinch defect defect to an external device.

[0078] The following processing may be performed without performing the low-magnification review process of step S1. For example, in the high-magnification review process of step S2, instead of steps S31 to S34, processing equivalent to steps S11 to S15 of the low-magnification review process is performed at a second magnification to identify the second detected defect D2. Then, the processing of steps S35 to S40 may be performed. In this case, the defect height measurement device 1 may not be provided with an optical microscope 51, and a white light interference microscope 52 may be used in the high-magnification review process.

[0079] If the review range setting unit 851 determines in steps S18 and S38 that the first and second review ranges A1 and A2 of the first and second range numbers have not been set, it may select the first and second detected defects D1 and D2 with the largest area from among the first and second detected defects D1 and D2 that have never been selected in the setting process of the first and second review ranges A1 and A2, regardless of whether they are included in the first and second review ranges A1 and A2 that have already been set (steps S16 and S36), and set the first and second review ranges A1 and A2 (steps S17 and S37). The review range setting unit 851 does not need to perform the process of step S35.

[0080] The height measurement unit 852 does not need to perform the process of step S58. The height measurement unit 852 may identify as measurement target defects Da all second detected defects D2 included in the second review range A2 whose area in the second range setting image P2 is equal to or greater than the second range threshold value, without performing the processing of steps S51 to S54. In step S51, the height measurement unit 852 may randomly select the second review range A2 regardless of the area of ​​the second detected defect D2. The low-magnification review process and the high-magnification review process may be performed using a white light interference microscope 52, or a laser microscope installed in place of the white light interference microscope 52, without installing an optical microscope 51 in the defect height measuring device 1. [Example]

[0081] Next, examples of the present invention will be described, but the present invention is not limited to these examples.

[0082] [Experimental Example 1] As Experimental Example 1, an experiment was conducted to investigate the correlation between the results of height measurement using the defect height measurement device 1 and the results of height measurement performed manually by an operator operating a laser microscope. Hereinafter, height measurement using the defect height measurement device 1 may be referred to as "automatic measurement," and height measurement performed manually by an operator operating a laser microscope may be referred to as "manual measurement."

[0083] First, 188 epitaxial wafers W, a defect height measuring device 1, and a laser microscope (tabletop laser microscope, manufactured by Keyence Corporation, model: VK-9510) were prepared. Then, the heights of the defects D were automatically measured for the three measurement target ranges A0 on the epitaxial wafer W by the measurement method shown in FIGS. 4 to 9 using the defect height measurement device 1. Furthermore, the height of defect D was manually measured for the three measurement ranges A0 on the automatically measured epitaxial wafer W. In the manual measurement, first, the epitaxial wafer W was placed on a mounting table. Then, defect D was searched for at 5x magnification without irradiating it with laser light, and the magnification was increased in order to 20x, 50x, and 150x to narrow down the defect D to be measured based on the area in the image. After this, the height of defect D was measured by raising the measurement unit of the laser microscope while irradiating laser light onto defect D narrowed down at 150x magnification.

[0084] Automatic and manual measurements were performed on 188 epitaxial wafers W, and the correlation between the maximum height measurement result obtained by automatic measurement and the maximum height measurement result obtained by manual measurement for each epitaxial wafer W was investigated, and the results shown in FIG. 11 were obtained. As shown in FIG. 11, the correlation coefficient was 0.956, confirming that there was a strong correlation between the automatic measurement and the manual measurement.

[0085] [Experimental Example 2] As Experimental Example 2, an experiment was conducted to compare the time required for height measurement by the above-mentioned automatic measurement with the time required for height measurement by the above-mentioned manual measurement.

[0086] First, as shown in Table 1, height measurements were performed automatically on different epitaxial wafers W over a period of 10 days. Furthermore, on the day when the automatic measurement was performed, manual measurement was performed on the epitaxial wafers W that had been automatically measured. The time required per day and the average time required for one epitaxial wafer W are shown in Table 1. Note that the required time does not include the time required to replace the epitaxial wafer W. As shown in Table 1, the average time required for automatic measurement was approximately 16.5 minutes, which was approximately 25 minutes shorter than manual measurement.

[0087] [Table 1] [Explanation of symbols]

[0088] 1...defect height measuring device, 5...observation unit, 51...optical microscope, 52...white light interference microscope (height measuring microscope), 6...moving unit, 8...control unit, 85...control unit, 851...review range setting unit, 852...height measuring unit, A0...measurement target range, A1, A2...first and second review ranges, D...defect, D1, D2...first and second detected defect, Da...measurement target defect, P1, P2...first and second range setting images, P3...measurement image, W...epitaxial wafer.

Claims

1. A method for measuring defect heights of an epitaxial wafer, the method being performed by a control device controlling a microscope, comprising: The control device a review range setting step of setting a review range that includes defects having an area equal to or larger than a range threshold value in a measurement target range of the epitaxial wafer captured by the microscope and is smaller than the measurement target range; a height measurement step of measuring the height of the defect based on a measurement image of the review range captured by the microscope at a magnification higher than that of the range setting image.

2. 2. The epitaxial wafer defect height measuring method according to claim 1, The review range setting step includes: a low-magnification review process for setting a first review range that includes defects whose area in a first range setting image captured at a first magnification of the measurement target range is equal to or larger than a first range threshold and is smaller than the measurement target range; a high-magnification review process for setting a second review range that is smaller than the first review range and that includes defects whose area in a second range setting image of the first review range is equal to or larger than a second range threshold, the second review range being captured at a second magnification higher than the first magnification; A method for measuring defect heights in an epitaxial wafer, wherein the height measurement step measures the height of the defect based on the measurement image of the second review range captured at a third magnification higher than the second magnification.

3. 3. The epitaxial wafer defect height measuring method according to claim 2, The low magnification review step includes: a first setting step of setting the first review range to include the defect with the largest area among the defects whose areas in the first range setting image are equal to or greater than the first range threshold; a second setting step of setting the first review range to include the defect whose area in the first range setting image is equal to or greater than the first range threshold and which has the largest area among the defects not included in the first review range that has already been set, the second setting step is repeated until a first number of first review ranges are set.

4. 4. The epitaxial wafer defect height measuring method according to claim 3, The high magnification review step includes: a third setting step of setting the second review range to include the defect with the largest area among the defects whose areas in the second range setting image are equal to or greater than the second range threshold; a fourth setting step of setting the second review range to include the defect whose area in the second range setting image is equal to or greater than the second range threshold and which has the largest area among the defects not included in the second review range that has already been set, the fourth setting step is repeated until a second number of second review ranges are set.

5. 5. The epitaxial wafer defect height measuring method according to claim 4, The high magnification review step includes: a defect height measurement method for an epitaxial wafer, further comprising an exclusion process of excluding the first review range in which the number of defects equal to or greater than the second range threshold included in the second range setting image is less than an exclusion reference value from the targets for setting the second review range in the third setting process and the fourth setting process.

6. The epitaxial wafer defect height measuring method according to any one of claims 2 to 5, The height measuring step includes: a measurement target specifying step of specifying the defect whose area in the measurement image is equal to or greater than a target threshold as a measurement target defect; a calculation step of calculating the height of the measurement target defect based on a plurality of the measurement images taken after the measurement target identification step from mutually different height positions relative to the epitaxial wafer.

7. The epitaxial wafer defect height measuring method according to any one of claims 2 to 5, the first range setting image and the second range setting image are captured by an optical microscope constituting the microscope, A method for measuring defect heights in an epitaxial wafer, wherein the measurement image is captured by a white light interference microscope or a laser microscope that constitutes the microscope.

8. An epitaxial wafer defect height measuring device, A microscope and a control device that controls the microscope to measure the height of defects in the epitaxial wafer; The control device a review range setting unit that sets a review range that includes defects whose area in a range setting image of a measurement target range of the epitaxial wafer captured by the microscope is equal to or larger than a range threshold and is smaller than the measurement target range; and a height measurement unit that measures the height of the defect based on a measurement image of the review range captured by the microscope at a magnification higher than that of the range setting image.

9. 9. The epitaxial wafer defect height measuring device according to claim 8, The review range setting unit a low-magnification review process for setting a first review range that includes defects whose area in a first range setting image captured at a first magnification of the measurement target range is equal to or larger than a first range threshold and is smaller than the measurement target range; a high-magnification review process for setting a second review range that is smaller than the first review range and that includes defects whose area in a second range setting image of the first review range is equal to or larger than a second range threshold, the second review range being captured at a second magnification higher than the first magnification; The height measurement unit measures the height of the defect based on the measurement image of the second review range captured at a third magnification higher than the second magnification.

10. 10. The epitaxial wafer defect height measuring device according to claim 9, the microscope is composed of a height measuring microscope which is a white light interference microscope or a laser microscope, and an optical microscope; the first range setting image and the second range setting image are captured by the optical microscope; The measurement image is captured by the height measuring microscope.

11. The epitaxial wafer defect height measuring device according to any one of claims 8 to 10, a moving unit that moves the microscope, The control device controls the microscope and the moving unit to measure the height of the defect in the epitaxial wafer.

Images