Method for measuring the wafer profile

Abstract

Method for measuring a wafer profile while holding an edge region of the wafer (W), using a flatness measuring system (1) in which the wafer (W) is arranged, and which comprises a first optical system (2) and a second optical system (3) arranged on a first principal surface and a second principal surface of the wafer (W), respectively, the method comprising: a first step of measuring both a surface variation on the first main surface and a surface variation on the second main surface of the wafer (W) using only any optical system (2, 3) of the first optical system (2) and of the second optical system (3); a second step of calculating an edge-region holding deformation amount caused by holding the edge region of the wafer (W) by using the surface variation on the first principal surface and the surface variation on the second principal surface measured with one of the optical systems (2, 3); and a third step of calculating an actual curvature value of the wafer (W) by subtracting the edge area holding deformation amount from a curvature value output by the flatness measuring system (1).

Description

TECHNICAL AREA

[0001] The present invention relates to a method for measuring a wafer profile using a wafer flatness measuring system of the edge-area holder type. STATE OF THE ART

[0002] As increasingly narrow design rules are developed, quality requirements now include not only the flatness of the wafer (thickness unevenness) but also the smoothness of the wafer profile (curvature (English: warp), SORI).

[0003] Generally, the curvature of a wafer is determined as waviness relative to the wafer's thickness centerline. In this case, the wafer's edge region is held by a specific method. With WaferSight and AFS, available from KLA Tencor, grippers are mechanically brought into direct contact at three points with beveled portions of the wafer towards its center to hold it. Depending on the gripper conditions and the type of force applied to these three points, an external force is exerted on the wafer, causing it to deform slightly and acquire a shape different from its original one.

[0004] JP 2007-64748 A states in paragraph 44 that the wafer deformation amounts vary between the different support positions due to the anisotropic influence of the elastic modulus of the silicon single crystal. However, patent document 1 does not indicate that a difference between the measuring systems plays a role in this case.

[0005] US 2005 / 0255610 A1 describes a method for evaluating the shape of a semiconductor wafer, comprising the following steps: measuring shape data of a semiconductor wafer by scanning a front and / or back surface of the semiconductor wafer; calculating a differential profile by a differential process of the measured shape data; analyzing the obtained differential profile and determining a surface property of the wafer; and evaluating the shape of the semiconductor wafer. US 2014 / 0268172 A1 describes a method for providing a measurement technique in a bright-field differential interference contrast system. The method may include generating a first and a second beam from a first light beam. The first and second beams have circular cross-sections and form first, partially overlapping sampling points that are radially offset on a substrate.A third and fourth beam are generated from the first light beam or a second light beam. The third and fourth beams have elliptical cross-sections and form second, partially overlapping scan points that are tangentially shifted on the substrate. At least a portion of the substrate can be scanned using the first and second partially overlapping scan points while the substrate is rotated. Radial and tangential inclinations can be determined from measurements obtained from scanning using the first and second partially overlapping scan points. These inclinations can be used to determine the wafer shape or any localized topographic features.

[0006] From DE 198 26 319 A1, a pair of optical profile measurement systems is known, which are provided at positions aligned on both sides of a wafer held vertically at its edge. Each system comprises a light source for emitting a measuring light beam, a collimator lens for focusing the light beam into a collimated beam, a plane glass plate for transmitting the collimated light beam, a light detector for receiving the light beams reflected by the collimator lens from a surface of the wafer and from a reference plane of the plane glass plate, and a computer for processing interference patterns that occur between the surface of the wafer and the reference plane of the plane glass plate.

[0007] DE 10 2004 054 566 A1 describes a method for flattening a semiconductor wafer, comprising the following steps: a) spatially dependent measurement of a parameter characterizing the semiconductor wafer in order to determine the spatially dependent value of this parameter over the entire surface of the semiconductor wafer, b) etching of this entire surface of the semiconductor wafer under the influence of an etching medium and simultaneous exposure of this entire surface, wherein the removal rate of the etching treatment depends on the light intensity at the surface of the semiconductor wafer and wherein the light intensity is spatially predetermined such that the differences in the spatially dependent values ​​of the parameter measured in step a) are reduced by the spatially dependent removal rate. SUMMARY OF THE INVENTIONAL PROBLEM

[0008] The deformation caused by holding the wafer edge area has, in principle, little effect on the thickness measurement. However, since curvature is a measurement parameter directly related to the wafer profile, such deformation affects the reproducibility of the measured values, especially as a difference between systems.

[0009] The difference in curvature between the systems can be expressed numerically. However, it is unknown how the holding state deforms a wafer. Furthermore, it is impossible to know the true value of the curvature of a wafer being measured. Therefore, there is no way to know which system is reliable.

[0010] Fig. shows the difference in curvature measurement between the systems, each of which is a WaferSight2+ manufactured by KLA Tencor. Fig. The graph shows a bar chart of the standard deviation of each of the 15 wafers (diameter: 300 mm) with different shapes, collected in various steps (DSP, CMP, EPW). The standard deviation was determined by measurement with the five measurement systems. It can be seen that the standard deviations differ between the wafers. Due to synergistic effects between the wafer profiles and the edge holding conditions of the measurement systems, the difference between the systems with respect to the curvature value, which indicates the surface variations on the wafer profile, is not always the same.

[0011] The present invention was made with regard to the problems mentioned above. One objective of the present invention is to provide a method for measuring a wafer profile in order to enable the measurement of the actual wafer curvature value using a flatness measurement system and to successfully obtain a curvature value that is minimally affected by differences between the systems. SOLUTION TO THE PROBLEM

[0012] To achieve this goal, the present invention provides a method for measuring a wafer profile while holding an edge region of the wafer by using a flatness measuring system in which the wafer is arranged and which includes a first optical system and a second optical system arranged on a first main surface and a second main surface of the wafer, respectively, wherein the method comprises: a first step of measuring a surface variation on the first main surface and a surface variation on the second main surface of the wafer using only one optical system of the first optical system and the second optical system; a second step of calculating an edge retention deformation amount caused by retention of the wafer's edge region, by using the surface variation on the first principal surface and the surface variation on the second principal surface, measured with one of the optical systems; and a third step in calculating an actual curvature value of the wafer by subtracting the edge area holding deformation amount from a curvature value output by the flatness measurement system.

[0013] Previously, a flatness measurement system with a first optical system and a second optical system, as described above, was used to measure and output a curvature value while the edge region of the wafer was held. This curvature value included the deformation caused by holding the edge region, so the actual curvature value of the wafer was not obtained. In contrast, the inventive method for measuring a wafer profile is able to capture an actual wafer curvature value without the deformation caused by holding the edge region. In this way, it is possible to capture a curvature value that is hardly affected by differences between the systems and exhibits high reproducibility.

[0014] In this case, the first step will be the wafer is placed in the flatness measurement system to measure the surface variation on the first main surface using only one of the optical systems, and The surface variation on the second main surface can be measured with only one of the optical systems, so the wafer is placed in the flatness measurement system while being flipped over relative to a state of the wafer that was placed to measure the surface variation on the first main surface, and The second step may include the following: a stage of detecting an initial thickness irregularity of the wafer; a stage in detecting a second thickness irregularity of the wafer; and a stage of recording the edge area holding deformation amount, wherein at the stage of detecting an initial thickness irregularity of the wafer, The surface variation measured in the first step is corrected on the second main surface in order to capture a corrected surface variation on the second main surface, and The first thickness irregularity is detected from a difference between the surface variation measured in the first step on the first main surface and the corrected surface variation on the second main surface. at the stage of detecting a second thickness irregularity of the wafer, the second thickness irregularity is detected from a difference between the surface variation measured with one of the optical systems on the first main surface and a surface variation measured with the other optical system on the second main surface, and at the stage of recording the edge area holding deformation amount, An average value of the first thickness irregularity and an average value of the second thickness irregularity are calculated. a difference between the average value of the second thickness irregularity and the average value of the first thickness irregularity is added to the first thickness irregularity to capture a corrected first thickness irregularity, and a value that is 1 / 2 of the difference between the corrected first thickness unevenness and the second thickness unevenness, as the edge area holding deformation amount is set.

[0015] In this way, an actual wafer curvature value can be determined more reliably without taking into account the edge area deformation amount.

[0016] In this case, the wafer profile can be measured using several flatness measurement systems.

[0017] Even when multiple flatness measurement systems are used for the measurements described above, each measurement successfully obtains an actual wafer curvature value. This makes it possible to suppress large system differences between the measured curvature values. ADVANTAGEOUS EFFECTS OF THE INVENTION

[0018] As described above, the inventive method for measuring a wafer profile enables the elimination of the deformation caused by holding the wafer edge region during measurement with a flatness measuring system and the determination of an actual wafer curvature value. In this way, a curvature value can be determined without system differences and with high reproducibility. BRIEF DESCRIPTION OF THE DRAWINGS Fig. is a flowchart to illustrate an example of a method for measuring a wafer profile according to the present invention. Fig. This is an explanatory diagram showing the sequence of procedures 1 to 4. Fig. is an explanatory diagram showing the process of procedure 5. Fig. is an explanatory diagram showing the sequence of procedures 6 and 7. Fig. Figure 1 is a schematic view showing an example of a flatness measurement system that can be used in the inventive method for measuring a wafer profile. Fig. is an explanatory view showing an example of a mechanism for holding the edge region of a wafer during measurement. Fig. shows measurement images illustrating the deformation amounts measured at the edge area using five flatness measurement systems. Fig. is a diagram showing the standard deviations of the curvature (warp) bf values ​​measured with the five flatness measurement systems according to the measurement method according to the invention (the curvature bf value is a difference between the maximum and minimum value of the deviation from a best-fit plane of a wafer mid-plane (thickness mid-line) (i.e., the best-fit plane is calculated from the mid-plane)). Fig. is a diagram showing the standard deviations of the curvature bf values ​​measured with the five flatness measurement systems using a conventional method. DESCRIPTION OF THE EXECUTION FORMS

[0019] Embodiments of the present invention are described below with reference to the drawings, but the present invention is not limited thereto.

[0020] Fig. Figure 1 shows a flatness measurement system that can be used in a method for measuring a wafer profile according to the present invention.

[0021] As in Fig. As shown, a flatness measurement system 1 comprises a first optical system 2 and a second optical system 3, each located on either side of a wafer W placed in the flatness measurement system 1, the wafer W having a first principal surface and a second principal surface. The first optical system 2 and the second optical system 3 each contain a measurement unit based on optical interferometry. The first optical system 2 is capable of measuring a distance between a sensor and the first principal surface of the wafer W and of measuring surface variation on the first principal surface. In contrast, the second optical system 3 is capable of measuring a distance between a sensor and the second principal surface and of measuring surface variation on the second principal surface.In this way, the surface variation on the first main surface and the surface variation on the second main surface can be measured separately.

[0022] Furthermore, the first optical system 2 and the second optical system 3 can measure a thickness unevenness (thickness distribution (flatness)) of the wafer W and a curvature value of the wafer W.

[0023] A CCD camera can capture the entire shape of a wafer, and it is possible to record an image of the entire surface of a wafer with a pixel size of 0.2 mm × 0.2 mm per point. For a wafer with a diameter of 300 mm, approximately 1.76 million data elements per layer of the wafer can be captured.

[0024] Fig. It further shows an example of a mechanism that holds the edge of a wafer during measurement.

[0025] As in Fig. As shown, the flatness measuring system 1, for example, has three grippers 4 to 6. These grippers 4 to 6 hold the edge region of a vertically oriented wafer, while the first optical system 2 and the second optical system 3 enable the measurement of the wafer W's profile. In this example, the tip is set to 0° in the vertical direction, and a notch is located counterclockwise at 270°. The wafer W (diameter: 300 mm, plane orientation: (100)) is held by the grippers 4 to 6 at three points: 0°, 135°, and 225°.

[0026] Note that in Fig. The DF1 to DF3 tools are designed to suppress vibrations of the wafer W and are each positioned between two of the three grippers 4 to 6. However, DF1 to DF3 have only minimal contact with the wafer W, and no active force is exerted on the wafer W. Unlike grippers 4 to 6, DF1 to DF3 are not holders.

[0027] Specific examples of the flatness measurement system 1 described above are Wafersight and AFS, offered by KLA Tencor, etc. It goes without saying that these are not the only examples.

[0028] The following describes the inventive method for measuring a wafer profile using the flatness measuring system 1 described above.

[0029] Fig. shows an example of the procedure of the measurement method according to the invention.

[0030] The method according to the invention comprises: a first step of measuring one surface variation each on the first main surface and on the second main surface of the wafer W using only any optical system of the first optical system 2 and the second optical system 3; a second step of calculating an edge retention deformation amount caused by holding the edge region of the wafer W, using the surface variations on the first and second main surfaces thus measured; and a third step of calculating an actual wafer curvature value by subtracting the edge retention deformation amount from a curvature value output by the flatness measuring system.

[0031] The significance of the first step will be described here. It is assumed that if a certain degree of deformation, attributable to the holding of the edge region (edge ​​region holding deformation amount), is present on wafer W, a standard measurement is performed according to a conventional measuring procedure (the surface variation on the first main surface of the wafer is measured with the first optical system 2, the surface variation on the second main surface of the wafer is measured with the second optical system 3). In this case, according to the method of thickness calculation, the influence of the profile change caused by the holding of the edge region of wafer W does not appear in the parameter (wafer thickness unevenness W) obtained from the measured surface variation.This is because, in the aforementioned ordinary measurement, when the front and back surfaces (the first main surface and the second main surface) of the wafer are measured with different optical systems (the first optical system and the second optical system), the amount of deformation in the wafer profile due to the holding scheme and the original waviness on the wafer profile are canceled out by calculating the difference between the front and back surfaces and cannot be captured by the conventional measurement method.

[0032] In contrast, when measuring the first and second main surfaces of the wafer with only one of the optical systems, as in the present invention, the influence of wafer deformation due to wafer holding is included and remains in the parameter (thickness unevenness (thickness distribution)). Thus, the edge holding deformation amount can be extracted from the parameter and subtracted from a curvature value (raw curvature value) measured by the measuring system. In this way, the inventor has found that an actual wafer curvature value can be easily determined with minimal influence from the edge holding deformation amount.

[0033] Each step is described in detail below. <First step>

[0034] As mentioned previously, only one of the optical systems of the first optical system 2 and the second optical system 3 is used at a time to measure the surface variation on the first main surface and the surface variation on the second main surface of the wafer W. It is also possible to use only the second optical system 3 for the measurement, although in the present example only the first optical system 2 is used. This measurement procedure is described below as an example.

[0035] Fig. is an explanatory diagram that shows the sequence of the following procedures 1 to 4.

[0036] First, a wafer is placed in the flatness measurement system to measure the surface variation on the first main surface using only one of the optical systems (Method 1).

[0037] The surface variation of the first main surface of the wafer W is measured using the first optical system 2.

[0038] Next, the surface variation on the second main surface is measured using only one of the optical systems, such that the wafer is placed in the flatness measurement system while being flipped over with respect to a state of the wafer that was placed to measure the surface variation on the first main surface (Method 2).

[0039] Here, the same wafer W is flipped, and the surface variation of the second primary surface of wafer W is measured using the first optical system 2. As described above, wafer W is flipped from the state of wafer W that was positioned to measure the surface variation on the first primary surface. In this flipped state, wafer W is positioned and measured using the first optical system 2, just as in the case of the first primary surface. < Second step>

[0040] The second step roughly comprises: a stage of detecting a first thickness irregularity of the wafer; a stage of detecting a second thickness irregularity of the wafer; and a stage of detecting an edge area holding deformation amount. (Stage of detecting the first thickness irregularity)

[0041] The surface variation measured in the first step on the second main surface is corrected to obtain a corrected surface variation on the second main surface (Method 3).

[0042] First, the surface variation of the second primary surface of wafer W, measured in the first step, is plotted in an r-theta coordinate system, and any deviation of the r-theta coordinates due to the rotation of wafer W in the first step is corrected. Because the wafer is rotated, the phase of the surface variation of the second primary surface in the r-theta coordinate system differs from that of the first primary surface. Therefore, the phase is corrected. Specifically, the phase is adjusted so that the surface variation of the second primary surface is positioned relative to the side of the first primary surface. More precisely, the surface variation of the second primary surface measured in the first step is rotated around an axis of rotation that corresponds to the axis of rotation during the measurement of the second primary surface in the first step.The rotations in the first step and here can be carried out based on a notch in the wafer W or similar.

[0043] In addition, a correction is performed to reverse the sign of the surface variation.

[0044] This allows the ratio of surface variation on the second main surface to be adjusted in the r-theta coordinates and the state of surface variation as seen from the first main surface side of the wafer (corrected surface variation on the second main surface).

[0045] Subsequently, the first thickness irregularity is determined from a difference between the surface variation measured in the first step on the first main surface and the corrected surface variation on the second main surface (Method 4).

[0046] In principle, as described for the following method 5, thickness unevenness of the wafer W is detected by subtracting (the surface variation on the first main surface measured by the first optical system 2) from (the surface variation on the second main surface measured by the second optical system 3) according to conventional measurement. In contrast, the subtraction of (the surface variation on the first main surface measured in the first step) from (the corrected surface variation on the second main surface) according to the present invention is merely a subtraction between the surface variations detected only by the first optical system 2. Accordingly, a preliminary thickness unevenness is calculated that contains no information about the absolute thickness. (Stage of detecting the second thickness irregularity)

[0047] The second thickness irregularity is then determined from a difference between the surface variation measured with one of the optical systems on the first main surface and a surface variation measured with the other optical system on the second main surface (Method 5). Fig. is an explanatory diagram showing the process of procedure 5.

[0048] As described above, the first thickness variation cannot be described as an actual wafer thickness variation. Therefore, in order to detect a thickness variation that is only measured with the first optical system 2 in the procedure 6 described later, a thickness variation of the wafer (second thickness variation) is first measured here according to the usual measurement procedure using both the first optical system 2 and the second optical system 3 by a standard procedure (conventional ordinary measurement procedure) of the measurement system itself. In particular, the surface variation on the second main surface, measured with the second optical system 3, is subtracted from the surface variation on the first main surface, measured with the first optical system 2, in order to detect the second thickness variation.

[0049] It should be noted that the measurements of surface variation on the first main surface with the first optical system 2 and the surface variation on the second main surface with the second optical system 3 can be carried out in this case, or that they can be carried out beforehand to detect these surface variations in advance. These measurements can be carried out, for example, in procedure 1. (Stage of recording the edge area holding deformation amount)

[0050] Fig. is an explanatory diagram showing the sequence of the following procedures 6, 7.

[0051] An average value of the first thickness unevenness and an average value of the second thickness unevenness are calculated, and a difference between the average value of the second thickness unevenness and the average value of the first thickness unevenness is added to the first thickness unevenness to capture a corrected first thickness unevenness (Procedure 6).

[0052] As described in the section on Procedure 5, this Procedure 6 captures a thickness unevenness measured only by the first optical system 2. First, an average value for the first thickness unevenness and an average value for the second thickness unevenness are calculated. For example, if the pixel size per digit is 0.2 mm × 0.2 mm, as described above, each of the first and second thickness unevennesses has a collection of approximately 1.76 million data elements for the thickness per layer of a wafer, assuming the wafer has a diameter of 300 mm. The average value of the first thickness unevenness and the average value of the second thickness unevenness each represent an average of such data.

[0053] In order to ensure that the average value of the first thickness irregularity (average thickness) is equal to the average value of the second thickness irregularity (average thickness), a value obtained by subtracting the average value of the first thickness irregularity from the average value of the second thickness irregularity is added to the data on the entire plane of the first thickness irregularity in order to capture the thickness irregularity measured only with the first optical system 2 (corrected first thickness irregularity), the average value of which is equal to that of the second thickness irregularity.

[0054] Next, a value corresponding to 1 / 2 of the difference between the corrected first thickness unevenness and the second thickness unevenness is set as the edge area holding deformation amount (Procedure 7).

[0055] In this way, the second thickness irregularity is subtracted from the corrected first thickness irregularity, and the difference is multiplied by 1 / 2. This allows only the amount of deformation attributable to holding at the edge region to be extracted and quantified. <Schritt 3>

[0056] The actual curvature value of the wafer is calculated by subtracting the edge retention deformation amount from a curvature value output by the flatness measurement system (Method 8).

[0057] In this way, the calculated edge-hold deformation amount is subtracted from a raw curvature value (including wafer deformation due to edge-holding) directly output by flatness measurement system 1. Curvature is an indicator of the maximum difference between the thickness centerline in a wafer cross-section and a best-fit plane of the entire wafer, or a best-fit plane calculated from three predefined points. The raw curvature value from flatness measurement system 1 is automatically calculated based on the state that includes the influence of edge-hold deformation. To eliminate this additional influence, the edge-hold deformation amount is subtracted from the raw curvature value as described in Procedure 8. This allows a curvature value (from which the edge-hold deformation amount is eliminated) to be obtained that is approximately true.In this way, a difference between the measuring systems with regard to the curvature value can be limited, whereby the difference is caused by the edge area holding deformation.

[0058] It should be noted that the specific procedures in steps 1 and 2 are not limited. Nevertheless, the substantially true curvature value can still be reliably determined if the boundary deformation amount is determined by procedures including those 1 to 7.

[0059] When multiple flatness measurement systems are used, the deformation amounts attributable to edge-holding deformation by grippers in the measurement systems are typically reflected as differences between the systems. In contrast, the present invention, as described above, makes it possible to eliminate the edge-holding deformation amount that causes such a difference between the systems and can thus improve and reduce the difference between the systems. Accordingly, even when using multiple measurement systems, it is possible to suppress a large variation in curvature values ​​between the measurement systems. EXAMPLE

[0060] The present invention will now be described in particular with reference to an example and a comparative example. However, the present invention is not limited to these. (Comparative example)

[0061] Using several (here five) flatness measurement systems (#A to #E), the curvature values ​​of 15 wafers, each with a diameter of 300 mm, originating from DSP, CMP, and EPW processes, were measured according to the conventional standard measurement method. Consequently, the result obtained was similar to that in Fig. As shown in this diagram of the standard deviations of the curvature values ​​due to the differences between the systems, the standard deviations varied for all wafers due to the synergistic effect of the wafer profiles and the edge retention conditions. The maximum standard deviation of the curvature value due to the differences between the systems was approximately 0.8 µm. (Example)

[0062] To verify the effectiveness of the present invention, the same 15 wafers and the same five flatness measurement systems were used as in the case of Fig. for the measurements according to the in Fig. The illustrated procedure of the measurement method according to the invention is used.

[0063] Fig. The image shows measurement images illustrating the edge retention deformation values ​​recorded by the five flatness measurement systems. The "range" indicates the maximum difference between the various edge retention deformation values ​​within a wafer plane. It is evident that system #A exhibits the smallest edge retention deformation value (range = 389.7 µm) and system #B the largest edge retention deformation value (range = 2470.2 µm).

[0064] These edge retention deformation amounts were calculated for all wafers and measuring systems and subtracted from the raw curvature values ​​of the measuring systems to calculate the true curvature values.

[0065] Fig. This shows the standard deviations of the calculated true curvature values ​​of the wafers between the measurement systems. Notably, to facilitate comparison with the results of the conventional method, the graph bars of... Fig. The graph bars on the left are from the conventional method, while the graph bars on the right are from the present invention.

[0066] The measurement method according to the invention reduced and improved the standard deviations of the curvature values ​​between the measurement systems, which clearly demonstrates the effect of reducing the differences between the systems.

Claims

[1] Method for measuring a wafer profile while holding an edge region of the wafer (W) using a flatness measuring system (1) in which the wafer (W) is arranged and which comprises a first optical system (2) and a second optical system (3) arranged on a first principal surface and a second principal surface of the wafer (W), respectively, the method comprising: a first step of measuring both a surface variation on the first main surface and a surface variation on the second main surface of the wafer (W) using only any optical system (2, 3) of the first optical system (2) and of the second optical system (3); a second step of calculating an edge-region holding deformation amount caused by holding the edge region of the wafer (W) by using the surface variation on the first principal surface and the surface variation on the second principal surface measured with one of the optical systems (2, 3); and a third step of calculating an actual curvature value of the wafer (W) by subtracting the edge area holding deformation amount from a curvature value output by the flatness measuring system (1). [2] Method for measuring a wafer profile according to claim 1, wherein In the first step, the wafer (W) is placed in the flatness measurement system (1) to measure the surface variation on the first main surface using only one of the optical systems (2, 3), and the surface variation on the second main surface is measured with only one of the optical systems (2, 3), so that the wafer (W) is placed in the flatness measurement system (1) while it is flipped over with respect to a state of the wafer (W) that was placed to measure the surface variation on the first main surface, and The second step includes: a stage of detecting an initial thickness irregularity of the wafer (W); a stage of detecting a second thickness irregularity of the wafer (W); and a stage of recording the edge area holding deformation amount, wherein at the stage of detecting an initial thickness irregularity of the wafer (W), The surface variation measured in the first step is corrected on the second main surface in order to capture a corrected surface variation on the second main surface, and The first thickness irregularity is detected from a difference between the surface variation measured in the first step on the first main surface and the corrected surface variation on the second main surface. at the stage of detecting a second thickness irregularity of the wafer (W), the second thickness irregularity is determined from a difference between the surface variation measured with one of the optical systems (2, 3) on the first main surface and a surface variation measured with the other optical system (2, 3) on the second main surface, and at the stage of recording the edge area holding deformation, An average value of the first thickness irregularity and an average value of the second thickness irregularity are calculated. a difference between the average value of the second thickness irregularity and the average value of the first thickness irregularity is added to the first thickness irregularity to capture a corrected first thickness irregularity, and a value that is 1 / 2 of a difference between the corrected first thickness irregularity and the second thickness irregularity is set to the edge area holding deformation amount. [3] Method for measuring a wafer profile according to claim 1 or 2, wherein the wafer profile is measured using a plurality of flatness measurement systems (1).

Images