Compensation methods for wafer morphology detection data

Abstract

This application provides a compensation method for wafer morphology detection data, relating to the field of wafer morphology detection. The method includes: for each scanning trajectory, determining a gravity compensation value based on the crystal orientation angle; performing gravity compensation on the original measurement data s(x) of the scanning trajectory using the gravity compensation value to obtain gravity-compensated first compensation data for the scanning trajectory; determining a tilt compensation value for each scanning trajectory; for each scanning trajectory, performing tilt compensation on the first compensation data of the scanning trajectory using the tilt compensation value to obtain tilt-compensated second compensation data for the scanning trajectory; determining the warpage of the scanning trajectory using the second compensation data of the scanning trajectory; and determining the warpage of the wafer under test based on the warpage of multiple scanning trajectories. This method can accurately and efficiently detect the surface morphology of wafers.

Description

Technical Field

[0001] This application relates to the field of wafer morphology detection, and in particular to a method for compensating wafer morphology detection data. Background Technology

[0002] With the rapid development of semiconductor technology, high-precision inspection of wafer surface morphology has become an indispensable part of semiconductor manufacturing. The wafer placement method directly affects process stability and inspection accuracy. The mainstream wafer placement methods are horizontal placement or vertical clamping, while morphology inspection involves ranging sensors and other optical inspection systems.

[0003] Currently, the wafer clamping methods used in related technologies can affect the wafer's warpage, and improper clamping increases the risk of breakage. Furthermore, these methods involve complex equipment. Optical inspection systems, such as interferometers, have stringent environmental requirements, often operating in Class 1000 cleanrooms. Their small measurement range makes them unsuitable for large wafers, and they are also time-consuming and costly. Horizontally placed wafers are subject to gravity, causing warpage changes, requiring data compensation for warpage during parameter calculations. Two relevant compensation methods exist: wafer flipping and theoretical modeling. However, wafer flipping requires two measurements of the same wafer, which is time-consuming; theoretical modeling only provides a compensation formula based on center-point support, which cannot meet current support conditions. Therefore, these compensation methods cannot accurately and efficiently detect the surface morphology of wafers. Summary of the Invention

[0004] Therefore, it is necessary to provide a compensation method for wafer morphology detection data to address the above-mentioned technical problems. This method can accurately and efficiently detect the surface morphology of wafers.

[0005] In a first aspect, this application provides a method for compensating wafer morphology detection data, the method comprising:

[0006] Obtain raw measurement data s(x) of multiple scan trajectories of the wafer under test; where x is the position coordinate along the scan trajectory, and the raw measurement data s(x) is the height value of the surface of the wafer under test from the reference plane at each position coordinate on the scan trajectory;

[0007] For each of the scanning trajectories, the crystal orientation angle between the scanning trajectory and the wafer under test is obtained, and the gravity compensation value under the crystal orientation angle is determined based on the crystal orientation angle; for each of the scanning trajectories, the original measurement data s(x) of the scanning trajectory is gravity compensated using the gravity compensation value to obtain the first compensated data of the scanning trajectory after gravity compensation;

[0008] Using the first compensation data of multiple scan trajectories, a tilt compensation value for each scan trajectory is determined; for each scan trajectory, tilt compensation is performed on the first compensation data of the scan trajectory using the tilt compensation value to obtain the tilt-compensated second compensation data of the scan trajectory.

[0009] For each scan trajectory, the warpage of that scan trajectory is determined using the second compensation data of the scan trajectory, and the warpage of the wafer under test is determined based on the warpage of multiple scan trajectories; wherein, the warpage is the difference between the maximum height value and the minimum height value of the wafer under test.

[0010] In one embodiment, determining the gravity compensation value at the crystal orientation angle includes:

[0011] Based on the crystal orientation angle, a first adjacent angle and a second adjacent angle are determined, and a first gravity compensation value at the first adjacent angle and a second gravity compensation value at the second adjacent angle are calculated; wherein, one of the first adjacent angle and the second adjacent angle is greater than the included angle and the other adjacent angle is less than the included angle;

[0012] Determine the first difference coefficient corresponding to the first adjacent angle and the second difference coefficient corresponding to the second adjacent angle;

[0013] The gravity compensation value is determined based on the first gravity compensation value and the first difference coefficient, the second gravity compensation value and the second difference coefficient, and the measured thickness value of the wafer to be tested.

[0014] In one embodiment, a first adjacent angle and a second adjacent angle are determined based on the crystal orientation angle, and a first gravity compensation value at the first adjacent angle and a second gravity compensation value at the second adjacent angle are calculated, including:

[0015] Based on the included angle, a first adjacent angle and a second adjacent angle are obtained from the pre-stored dataset, along with a first set of fitting coefficients corresponding to the first adjacent angle and a second set of fitting coefficients corresponding to the second adjacent angle.

[0016] The first gravity compensation value at the first adjacent angle is determined based on the first set of fitting coefficients and the fitting formula, and the second gravity compensation value at the second adjacent angle is determined based on the second set of fitting coefficients and the fitting formula.

[0017] In one embodiment, the generation of the pre-stored dataset includes:

[0018] For the target wafer, the scanning trajectory deformation curves under different target crystal orientation angles are obtained by traversing the target crystal orientation angles from 0° to 90° according to the target step size.

[0019] The deformation curve is fitted with an nth-order polynomial to obtain the fitting coefficients for each set of values ​​corresponding to different target crystal orientation angles; where n is an integer and n≤5;

[0020] The fitting coefficients corresponding to different target crystal orientation angles are stored to obtain the pre-stored dataset.

[0021] In one embodiment, determining the gravity compensation value based on the first gravity compensation value and the first difference coefficient, the second gravity compensation value and the second difference coefficient, and the measured thickness value of the wafer under test includes:

[0022] Determine the first product of the first gravity compensation value and the first difference coefficient, the second product of the second gravity compensation value and the second difference coefficient, and calculate the sum of the products of the first product and the second product;

[0023] Determine the thickness ratio between the target wafer's thickness and the measured thickness of the wafer under test, and determine the gravity compensation value based on the thickness ratio and the sum of their products.

[0024] In one embodiment, the tilt compensation value for each of the multiple scan trajectories is determined using first compensation data, including:

[0025] For each of the aforementioned scanning trajectories, a first equation is established between the tilt compensation value, the first compensation data, and the true value to be determined on that scanning trajectory, resulting in a first set of equations; wherein, the first set of equations includes multiple first equations corresponding to multiple scanning trajectories;

[0026] The target distance between support points and the distance L from the support point to the center of the wafer under test are obtained. For each scan trajectory, a second equation is established between the tilt compensation value, the target distance, the distance L, and the true value to be determined on the scan trajectory, resulting in a second set of equations. The second set of equations includes multiple second equations corresponding to multiple scan trajectories. The support points are used to support the wafer under test.

[0027] Based on the first set of equations and the second set of equations, and using the least squares method, the tilt compensation value for each scan trajectory is determined.

[0028] In one embodiment, determining the tilt compensation value for each scan trajectory based on the first set of equations and the second set of equations using the least squares method includes: establishing constraints on the first set of equations and the second set of equations based on the true values ​​to be determined on at least two scan trajectories to obtain a joint set of equations; and applying the least squares method to the joint set of equations to obtain the tilt compensation value for each scan trajectory.

[0029] For each of the scanning trajectories, tilt compensation is performed on the first compensation data of the scanning trajectory using the tilt compensation value to obtain the second compensation data of the scanning trajectory after tilt compensation. This includes: for each scanning trajectory, substituting the tilt compensation value of the scanning trajectory and the first compensation data into the first equation corresponding to the scanning trajectory to obtain the second compensation data of the scanning trajectory after tilt compensation.

[0030] In one embodiment, there are four scanning trajectories. The current scanning trajectory is obtained by scanning the wafer under test after rotating the crystal orientation angle corresponding to the previous scanning trajectory by 45 degrees.

[0031] The second set of equations comprises four second equations, and the combined set of equations comprises six equations.

[0032] Secondly, embodiments of this application provide a wafer morphology measuring device, the measuring device comprising:

[0033] Lateral displacement component, used for horizontal movement;

[0034] A rotary lifting assembly is used to rotate and / or lift a wafer under test. The rotary lifting assembly rotates and / or lifts the wafer under test to rotate it by a corresponding angle. The rotary lifting assembly includes a first rotary lifting member and a second rotary lifting member.

[0035] The upper and lower probe assembly includes two ranging sensors, one upper and one lower.

[0036] A wafer placement stage is used to place the wafer under test, and it is configured with a three-point support structure. The lateral displacement component drives the wafer placement stage to move the wafer under test horizontally between the first rotating lifting member and the second rotating lifting member. When the wafer under test passes the upper and lower distance sensors, the upper and lower distance sensors collect the height value of the wafer under test surface from the reference plane in the diameter direction to obtain the original measurement data s(x) of multiple scanning trajectories of the wafer under test.

[0037] A control module, which stores a computer program, implements the compensation method for wafer morphology detection data according to any one of the first aspects when the computer program is executed.

[0038] In one embodiment, the measuring device further includes a five-axis adjustment mechanism for adjusting the X / Y axis translation, Z axis lifting, and pitch yaw of the two ranging sensors.

[0039] The five-axis adjustment mechanism includes a Y-axis direction adjustment module for translating the two range sensors along the Y-axis, an X-axis direction adjustment module for translating the two range sensors along the X-axis, a Z-axis lifting module for raising and lowering the two range sensors along the Z-axis, and a pitch yaw adjustment module for adjusting the pitch and yaw of the two range sensors.

[0040] The aforementioned compensation method and measuring device for wafer morphology detection data only require the use of multiple scanning trajectories of the wafer under test. First, gravity compensation is performed on the original measurement data s(x) of the multiple scanning trajectories to correct the deformation of the wafer material itself under anisotropic gravity. This can efficiently and accurately subtract the different deformations caused by gravity in different crystal orientations, restoring the intrinsic morphology of the wafer under test in the absence of gravity. Furthermore, after gravity compensation, tilt compensation is performed to correct the error introduced by the equipment support, effectively eliminating the overall tilt error introduced by the small height difference between the three support points. This solves the error problem of the support system itself, making the warpage data obtained by the user closer to the true properties of the wafer, rather than the data distorted by the measuring fixture, gravity, and other factors. Attached Figure Description

[0041] Figure 1 This is a structural diagram of a wafer morphology measurement device in one embodiment;

[0042] Figure 2 This is a structural diagram of a three-point support structure in one embodiment;

[0043] Figure 3 This is a structural diagram of a five-axis adjustment mechanism in one embodiment;

[0044] Figure 4 A flowchart of a method for compensating wafer morphology detection data in one embodiment;

[0045] Figure 5 This is a schematic diagram of the scan trajectory in one embodiment;

[0046] Figure 6 This is a flowchart illustrating the determination of gravity compensation values ​​in one embodiment;

[0047] Figure 7 This is a flowchart for determining the tilt compensation value in one embodiment;

[0048] Figure 8 This is a schematic diagram showing the position markings of the support points on the scanning trajectory in one embodiment. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0050] In one embodiment, a wafer morphology measurement device is provided, such as... Figure 1 As shown, the measuring device includes a lateral displacement assembly 1, a rotary lifting assembly 2, an upper and lower probe assembly 3, an edge-finding assembly 4, a wafer placement stage 5, and a calibration wafer placement stage 6. The lateral displacement assembly 1 is capable of horizontal movement; the rotary lifting assembly 2 is used to rotate and / or raise / lower the wafer under test, rotating the wafer under test by a corresponding angle; the rotary lifting assembly 2 includes a first rotary lifting member 201 and a second rotary lifting member 202. The upper and lower probe assembly 3 includes two distance sensors; the wafer placement stage 5 is used to place the wafer under test and is configured with a three-point support structure.

[0051] The lateral displacement component 1 drives the wafer placement stage 5 to move the wafer under test horizontally between the first rotating lifting component 201 and the second rotating lifting component 202. When the wafer under test passes the upper and lower distance sensors, the upper distance sensor measures the height of the upper surface and the lower distance sensor measures the height of the lower surface. The local thickness of the wafer under test is calculated. The upper and lower distance sensors collect the height values ​​of the lower surface to the upper surface of the wafer under test in the diameter direction to obtain the original measurement data s(x) of multiple scanning trajectories of the wafer under test.

[0052] The wafer morphology measuring device also includes a control module that stores a computer program. When the computer program is executed, it implements the wafer morphology detection data compensation method involved in the following embodiments.

[0053] Furthermore, the wafer placement stage 5 is connected to the calibration wafer placement stage 6 and can move with the lateral displacement assembly 1. The upper and lower probe assembly 3 measures the calibration wafer placed on the calibration wafer placement stage 6 and the wafer to be tested placed on the wafer placement stage 5. In addition, the three-point support structure at the wafer placement stage 5 can perform basic positioning of the three support points, and with fine-tuning screws, improves the adjustability of the support plane. Figure 2 As shown, the three-point support structure includes a positioning support rod 501, a three-point support positioning fixture 502, and a three-point support positioning fixture fine-tuning structure 503. For example, for a 12-inch wafer, the three points are located on a circle with a diameter of 200mm, and the three-point support positioning fixture can be finely adjusted simultaneously.

[0054] In addition, a calibration plate can be set. Before each measurement of the wafer to be tested, the upper and lower probe assemblies 3 measure the calibration plate. The calibration plate is fixed on the calibration plate placement stage 66. The wafer placement stage 5 and the calibration plate placement stage 6 are connected to each other and can be considered as one unit. The calibration plate can correct the wafer thickness under the current environment. After calibration, the influence of the surrounding environment (such as temperature and humidity factors) on the equipment can be reduced.

[0055] In one embodiment, such as Figure 3 As shown, the measuring device may also include a five-axis adjustment mechanism 8, which includes a Y-axis direction adjustment module 801 for translating the two range sensors along the Y-axis, an X-axis direction adjustment module 802 for translating the two range sensors along the X-axis, a Z-axis lifting module 803 for lifting the two range sensors along the Z-axis, and a pitch yaw adjustment module 804 for adjusting the pitch and yaw of the two range sensors.

[0056] Specifically, the five-axis adjustment mechanism 8 can adjust the two range sensors. The adjustment purposes are: 1) to adjust the distance between the two range sensors and the wafer, as different distances affect the measurement accuracy of the range sensors; 2) to adjust the perpendicularity of the two range sensors to the wafer, as different angles also affect the measurement accuracy of the range sensors; 3) to align the upper and lower range sensors, ensuring that the upper and lower range sensors measure the same area of ​​the wafer. The lateral direction of the range sensors can be adjusted by the Y-axis adjustment module 801 and the X-axis adjustment module 802, the height of the range sensors can be adjusted by the Z-axis lifting module 803, and the pitch and yaw adjustment module 804 can adjust the pitch and yaw angles of the range sensors.

[0057] In one embodiment, the rotary motor and linear motor in the rotary lifting assembly 2 can be replaced to meet higher precision requirements. Furthermore, the upper and lower ranging sensors can be replaced with different types of sensors to achieve wafer morphology detection at different process stages.

[0058] In one embodiment, the specific working principle of the wafer morphology measurement device can be as follows:

[0059] 1) The wafer to be tested is placed on the rotating lifting component 201 by hand or automated placement. The rotating lifting component 201 adsorbs the wafer to be tested by vacuum or other means. The edge finding component 4 determines the offset of the current position of the wafer to be tested based on the edge of the wafer to be tested.

[0060] 2) After the offset judgment is completed, the rotating lifting assembly 201 places the wafer under test onto the wafer placement stage 5 according to its own lifting mechanism. The transverse displacement assembly 1 moves the wafer placement stage 5 to correct the position offset of the wafer under test. Similarly, the edge finding assembly 4 can judge the position features on the wafer under test, such as V-grooves or flat edges.

[0061] 3) The wafer under test can be swapped between the wafer placement stage 5 and the edge-finding assembly 4 via the first rotating lifting member 201 or the second rotating lifting member 202. After edge finding is completed, the wafer under test moves on the wafer placement stage 5 following the lateral displacement assembly 1.

[0062] 4) The upper and lower probe assembly consists of two distance sensors, one on the top and one on the bottom. Each distance sensor can measure the distance from itself to the upper (or lower) surface of the wafer under test in real time. While the two distance sensors are measuring, the lateral displacement assembly 1 moves the wafer under test between the first rotating lifting member 201 and the second rotating lifting member 202. Each movement measures the data of the wafer under test in the diameter direction. This yields the raw measurement data s(x) of one scanning trajectory of the wafer under test.

[0063] 5) The lateral displacement component 1 moves from the first rotating lifting component 201 to the second rotating lifting component 202 (or from the second rotating lifting component 202 to the first rotating lifting component 201). The wafer under test can be rotated by a corresponding angle through the first rotating lifting component 201 (or the second rotating lifting component 202). The rotation angle can be set by the control module or the host computer to rotate the specified angle according to the set program.

[0064] According to the set procedure, repeat steps 3-5 until the wafer under test is measured. It should be noted that... Figure 1 As shown, the wafer under test is placed on three support points 502. The lateral displacement component 1 moves the wafer under test through the upper and lower probe components 3. After rotating the wafer under test by a corresponding angle, the lateral displacement component 1 moves the wafer under test between the first rotating lifting component 201 and the second rotating lifting component 202, thus measuring the data of the wafer under test in the diameter direction once, and obtaining the original measurement data s(x) of one scan trajectory of the wafer under test. Therefore, by repeatedly executing steps 3-5, the original measurement data s(x) of multiple scan trajectories of the wafer under test can be obtained.

[0065] In one embodiment, such as Figure 4 As shown, a method for compensating wafer morphology detection data is provided, which specifically includes the following steps 401 to 404:

[0066] Step 401: Obtain the original measurement data s(x) of multiple scanning trajectories of the wafer under test; where x is the position coordinate along the scanning trajectory, and the original measurement data s(x) is the height value of the wafer surface under test from the reference plane at each position coordinate on the scanning trajectory.

[0067] The measurement path of the upper and lower probe assemblies 3 on the wafer under test 7 is the scan trajectory 701. For example... Figure 5 As shown, the measurement path is within the diameter range. It should be noted that the original measurement data s(x) is the height value from the lower surface to the upper surface of the wafer under test at each coordinate on the scanning trajectory. This height value can be understood as the local thickness or net morphology of the wafer under test.

[0068] Step 402: For each scanning trajectory, obtain the crystal orientation angle between the scanning trajectory and the wafer 7 to be tested, and determine the gravity compensation value under the crystal orientation angle based on the crystal orientation angle; for each scanning trajectory, perform gravity compensation on the original measurement data s(x) of the scanning trajectory through the gravity compensation value to obtain the first compensation data of the scanning trajectory after gravity compensation.

[0069] like Figure 5 As shown, the wafer under test 7 includes a groove 702 representing the wafer crystal orientation 703. When the rotating lifting assembly 2 rotates the wafer under test 7, the relative position of the scanning trajectory 701 and the support point (i.e. the support point at the three-point support positioning fixture 502) does not change. There is an angle 704 between the scanning trajectory 701 and the wafer crystal orientation 703, which is the angle between the scanning trajectory and the crystal orientation of the wafer under test 7.

[0070] When the wafer under test (WUT) 7 is placed on three support points, its warpage will change due to gravity. Although the scanning path is fixed relative to the support points, the warpage caused by gravity will change after the wafer under test 7 is rotated due to the anisotropy of the material (general semiconductor materials are anisotropic, meaning that the same load will produce different wafer deformations under different crystal orientations). Therefore, data compensation for the amount of gravitational deformation under different crystal orientations is required.

[0071] Step 403: Using the first compensation data of multiple scanning trajectories, determine the tilt compensation value of each scanning trajectory; for each scanning trajectory, perform tilt compensation on the first compensation data of the scanning trajectory using the tilt compensation value to obtain the second compensation data of the scanning trajectory after tilt compensation.

[0072] Specifically, after gravity compensation is applied to the original measurement data s(x) of each scanning trajectory, the first compensation data of each scanning trajectory is obtained. Then, tilt compensation is applied to the first compensation data of each scanning trajectory to obtain the second compensation data of each scanning trajectory. That is, the data compensation of the wafer 7 under test includes gravity compensation and tilt compensation, and gravity compensation is applied before tilt compensation.

[0073] Step 404: For each scan trajectory, the warpage of the scan trajectory is determined using the second compensation data of the scan trajectory. The warpage of the wafer 7 under test is determined based on the warpage of multiple scan trajectories. The warpage is the difference between the maximum height value and the minimum height value of the wafer 7 under test.

[0074] Specifically, for each scan trajectory, the warpage corresponding to that scan trajectory can be understood as the difference between the maximum thickness (or maximum height) and the minimum thickness (or minimum height) of the wafer under test 7 along the scan trajectory. The final warpage of the wafer under test 7 can be the maximum warpage value among multiple scan trajectories, or it can be the average value. For example, assuming there are four scan trajectories, A, B, C, and D, and the warpage corresponding to trajectories C is calculated to be the largest, the warpage of trajectories C can be used as the warpage of the wafer under test 7. Alternatively, the average warpage values ​​of trajectories A, B, C, and D can be used as the warpage of the wafer under test 7.

[0075] In this application, multiple scanning trajectories of the wafer under test 7 are used to first perform gravity compensation on the original measurement data s(x) of the multiple scanning trajectories. This corrects the deformation of the wafer under test 7 material itself under anisotropic gravity, accurately subtracting the different deformation caused by gravity in different crystal orientations, and restoring the intrinsic morphology of the wafer under test 7 under zero gravity. Furthermore, after gravity compensation, tilt compensation is performed to correct the error introduced by the equipment support, effectively eliminating the overall tilt error introduced by the small height difference of the three support points. This solves the error problem of the support system itself, making the warpage data obtained by the user closer to the true properties of the wafer, rather than the data distorted by the measuring fixture, gravity, and other factors.

[0076] In one embodiment, such as Figure 6 As shown, step 402, which determines the gravity compensation value based on the crystal orientation angle, specifically includes the following steps 601 to 603:

[0077] Step 601: Determine the first adjacent angle and the second adjacent angle based on the crystal orientation angle, and calculate the first gravity compensation value under the first adjacent angle and the second gravity compensation value under the second adjacent angle; wherein, one of the first adjacent angle and the second adjacent angle is greater than the included angle and the other adjacent angle is less than the included angle.

[0078] Step 601 involves determining the first adjacent angle and the second adjacent angle based on the crystal orientation angle, and calculating the first gravity compensation value at the first adjacent angle and the second gravity compensation value at the second adjacent angle. Specifically, this includes:

[0079] The first adjacent angle and the second adjacent angle are obtained by matching the included angle from the pre-stored dataset, as well as the first set of fitting coefficients corresponding to the first adjacent angle and the second set of fitting coefficients corresponding to the second adjacent angle.

[0080] The first gravity compensation value at the first adjacent angle is determined based on the first set of fitting coefficients and fitting formulas, and the second gravity compensation value at the second adjacent angle is determined based on the second set of fitting coefficients and fitting formulas.

[0081] The generation of the pre-stored dataset specifically includes: for the target wafer, traversing the target crystal orientation angle from 0° to 90° according to the target step size to obtain the deformation curve of the scanning trajectory under different target crystal orientation angles; performing n-order polynomial fitting on the deformation curve to obtain the fitting coefficients corresponding to each set under different target crystal orientation angles; where n is an integer and n≤5; storing the fitting coefficients corresponding to each set under different target crystal orientation angles to obtain the pre-stored dataset.

[0082] Step 602: Determine the first difference coefficient corresponding to the first adjacent angle and the second difference coefficient corresponding to the second adjacent angle;

[0083] Step 603: Determine the gravity compensation value based on the first gravity compensation value and the first difference coefficient, the second gravity compensation value and the second difference coefficient, and the measured thickness value of the wafer 7 to be tested.

[0084] In step 603, the gravity compensation value is determined based on the first gravity compensation value and the first difference coefficient, the second gravity compensation value and the second difference coefficient, and the measured thickness value of the wafer 7 to be tested. Specifically, this includes:

[0085] Determine the first product of the first gravity compensation value and the first difference coefficient, the second product of the second gravity compensation value and the second difference coefficient, and calculate the sum of the products of the first product and the second product;

[0086] Determine the thickness ratio between the target wafer's thickness and the measured thickness of the wafer 7 to be tested. Based on the thickness ratio and the sum of their products, determine the gravity compensation value.

[0087] Specifically, gravity compensation requires first performing a series of modeling and simulations on the target wafer (the wafer used for modeling and simulation) to obtain the deformation curve at the scanning trajectory under the influence of gravity. Then, polynomial fitting is performed on the deformation curve to obtain the fitting coefficients. Since the size of the groove on the target wafer is much smaller than the size of the target wafer, the groove can be ignored during modeling. Because the wafer material is generally anisotropic, and the crystal orientation angles of the scanning trajectory are unknown before measurement, the simulation needs to start from 0° of crystal orientation A, traverse through crystal orientation B at 1° intervals (adjusting the value according to accuracy requirements) until returning to crystal orientation A, where A and B are the main crystal orientations of the target wafer in this crystal plane. Due to the symmetry of the crystal, the traversal range is less than 360°. Therefore, the pre-stored dataset contains polynomial fitting coefficients for the gravity deformation on the scanning trajectory of the target wafer under different crystal orientation angles.

[0088] For example, θ i θ is the crystal orientation angle between the i-th scan trajectory and the crystal orientation. i The first adjacent angle and the second adjacent angle are θ. j and θ j+1 For example, the crystal orientation angle θ i The first adjacent angle can be 28°, and the second adjacent angle can be 26°. For example, if the pre-stored dataset contains polynomial fitting coefficients for crystal orientation angles spaced 0.5° apart, such as polynomial fitting coefficients for crystal orientation angles of 26.5° and 27.5°, then the crystal orientation angle θ... i With a value of 27°, the first adjacent angle can be 26.5° and the second adjacent angle can be 27.5°.

[0089] Furthermore, based on the first set of fitting coefficients and fitting formulas, the first gravity compensation value at the first adjacent angle is determined, and based on the second set of fitting coefficients and fitting formulas, the second gravity compensation value at the second adjacent angle is determined. The specific fitting formulas are as follows:

[0090] ;

[0091] Among them, b0, b1, b2, b n Let be the fitting coefficient, and x be the coordinates (x-axis) of each position on the scanning trajectory. Using the specific fitting formula described above, the first gravity compensation value is obtained. and the first gravity compensation value .

[0092] Based on the crystal orientation angle θ i The first adjacent angle θ j and the second adjacent angle θ j+1 Calculate the first difference coefficient *a* corresponding to the first adjacent angle, and the second difference coefficient *1-a* corresponding to the second adjacent angle, where 0 ≤ *a* ≤ 1. Specifically:

[0093] ;

[0094] Based on the first gravity compensation value and the first difference coefficient, the second gravity compensation value and the second difference coefficient, and the measured thickness value of the wafer 7 to be tested, the gravity compensation value is determined. Specifically:

[0095] ;

[0096] Where t is the measured thickness of the wafer 7 to be tested, and t0 is the target wafer thickness value during modeling and simulation.

[0097] In this embodiment of the application (steps 601 to 603), the finite element method is used to simulate and calculate the gravity deformation of target wafers with different crystal orientations and thicknesses under three-point support. The fitting coefficients used for gravity compensation are extracted. Compared with the wafer flipping method, this method does not require flipping measurement of the wafer 7 under test, is less time-consuming and has less data that can be integrated into the software, and does not require the establishment of a database. Furthermore, it is more consistent with the actual support situation of the wafer 7 under test, and the gravity compensation is more accurate.

[0098] In one embodiment, such as Figure 7 As shown, step 403 uses the first compensation data of multiple scan trajectories to determine the tilt compensation value for each scan trajectory, specifically including the following steps 701 to 703:

[0099] Step 701: For each scan trajectory, establish a first equation between the tilt compensation value, the first compensation data, and the true value to be determined on the scan trajectory, to obtain a first set of equations; wherein, the first set of equations includes multiple first equations corresponding to multiple scan trajectories.

[0100] Step 702: Obtain the target distance between support points and the distance L from the support point to the center of the wafer 7 under test. For each scanning trajectory, establish a second equation between the tilt compensation value, target distance, distance L and the true value to be determined on the scanning trajectory to obtain a second set of equations. The second set of equations includes multiple second equations corresponding to multiple scanning trajectories. The support points are used to support the wafer 7 under test.

[0101] Step 703: Based on the first and second sets of equations and using the least squares method, determine the tilt compensation value for each scanning trajectory.

[0102] In step 703, the tilt compensation value for each scanning trajectory is determined based on the first and second sets of equations and using the least squares method. Specifically, this includes:

[0103] Based on the true values ​​to be determined on at least two scanning trajectories, constraints are established on the first and second sets of equations to obtain a joint set of equations; the least squares method is used on the joint set of equations to obtain the tilt compensation value for each scanning trajectory.

[0104] For each scanning trajectory, tilt compensation is performed on the first compensation data of the scanning trajectory using tilt compensation value to obtain the second compensation data of the scanning trajectory after tilt compensation. This includes: for each scanning trajectory, substituting the tilt compensation value of the scanning trajectory and the first compensation data into the first equation corresponding to the scanning trajectory to obtain the second compensation data of the scanning trajectory after tilt compensation.

[0105] Specifically, there are four scanning trajectories. The current scanning trajectory is obtained by scanning the wafer 7 under test after rotating the crystal orientation angle corresponding to the previous scanning trajectory by 45 degrees. The second set of equations includes four second equations, and the joint set of equations includes six equations.

[0106] For example, such as Figure 8 As shown, taking four-line scanning as an example, the scanning trajectory is the solid black line with arrows in the schematic diagram. The support points on the scanning trajectory are marked as 1-8 (markers 1 and 5 correspond to one scanning trajectory, marks 2 and 6 correspond to one scanning trajectory, marks 3 and 7 correspond to one scanning trajectory, and marks 4 and 8 correspond to one scanning trajectory). The support points for the scanning trajectories marked 1 and 5 are marked 3, 6, and 8. Let w be the true value to be determined at the mark, and s be the measured value at the mark. Considering only the effect of tilt, the following relationship exists, which is the first system of equations:

[0107] ;

[0108] ;

[0109] ;

[0110] ;

[0111] ;

[0112] ;

[0113] ;

[0114] Where k is the tilt of each scan trajectory at the marked point. Define Δ as the height difference between the support points (the two points affecting the slope of the scan trajectory, shown as support points at marks 6 and 8 in the diagram), 2d as the distance between the support points, and L as the distance from the support point to the center of the circle. From geometric relationships, the tilt amounts k1, k2, k3, and k4 can be obtained as follows:

[0115] ;

[0116] ;

[0117] ;

[0118] ;

[0119] The equations for the tilt amounts k1, k2, k3, and k4 are the second equations, and these four second equations constitute the second equation system.

[0120] By adding the constraints w1 = w5 and w3 = w7, we can obtain the joint system of equations:

[0121] ;

[0122] ;

[0123] ;

[0124] ;

[0125] ; ;

[0126] At this point, there are five unknowns and six equations. The tilt amounts k1, k2, k3, and k4 can be obtained by solving using the least squares method. For each scan trajectory, the first compensation data of the scan trajectory is tilted using the tilt compensation value to obtain the second compensation data after tilt compensation. Specifically, the second compensation data is the data after gravity compensation and tilt compensation in sequence. When performing tilt compensation, a mathematical model is established, representing the measured value (s) as a combination of the true topography value (w) and the tilt amount (k) caused by the height difference of the support points. A system of equations is constructed using data from multiple scan lines, and reasonable constraints are added (such as the topography values ​​of two relative points being equal). The optimal tilt amount k is then solved using the least squares method. The tilt plane formed by k is subtracted from the measured data, thereby eliminating the error caused by the unevenness of the support surface and improving the accuracy of the topography detection of the wafer under test.

[0127] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0128] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for compensating wafer morphology detection data, characterized in that, The method comprises: Obtaining original measurement data s(x) of a plurality of scanning tracks of a wafer to be measured; wherein x is a position coordinate along the scanning track, and the original measurement data s(x) is a height value of a surface of the wafer to be measured to a reference plane at each position coordinate on the scanning track; For each scanning track, an angle between the scanning track and a crystal direction of the wafer to be measured is obtained, and a gravity compensation value corresponding to the angle is determined based on the angle; for each scanning track, the original measurement data s(x) of the scanning track is compensated by the gravity compensation value to obtain first compensation data of the scanning track after gravity compensation; Using the first compensation data of a plurality of scanning tracks, a tilt compensation value of each scanning track is determined; for each scanning track, the first compensation data of the scanning track is compensated by the tilt compensation value to obtain second compensation data of the scanning track after tilt compensation; For each scanning track, a warping degree of the scanning track is determined using the second compensation data of the scanning track, and a warping degree of the wafer to be measured is determined based on the warping degrees of a plurality of scanning tracks; wherein the warping degree is a difference between a maximum height value and a minimum height value of the wafer to be measured; Determining the gravity compensation value corresponding to the angle based on the angle comprises: determining a first adjacent angle and a second adjacent angle based on the angle, and calculating a first gravity compensation value corresponding to the first adjacent angle and a second gravity compensation value corresponding to the second adjacent angle; wherein one of the first adjacent angle and the second adjacent angle is greater than the angle, and the other is less than the angle; determining a first difference coefficient corresponding to the first adjacent angle and a second difference coefficient corresponding to the second adjacent angle; and determining the gravity compensation value based on the first gravity compensation value and the first difference coefficient, the second gravity compensation value and the second difference coefficient, and a measured thickness value of the wafer to be measured.

2. The wafer topography detection data compensation method of claim 1, wherein, Determining the first adjacent angle and the second adjacent angle based on the angle, and calculating the first gravity compensation value corresponding to the first adjacent angle and the second gravity compensation value corresponding to the second adjacent angle, comprises: According to the angle, a first adjacent angle and a second adjacent angle are matched from a pre-stored data set, and a first set of fitting coefficients corresponding to the first adjacent angle and a second set of fitting coefficients corresponding to the second adjacent angle are matched; The first gravity compensation value corresponding to the first adjacent angle is determined based on the first set of fitting coefficients and a fitting formula, and the second gravity compensation value corresponding to the second adjacent angle is determined based on the second set of fitting coefficients and the fitting formula. 3.The wafer topography detection data compensation method of claim 2, wherein, The generation of the pre-stored data set comprises: For a target wafer, a target crystal direction angle is traversed from 0° to 90° according to a target step size to obtain a deformation curve of a scanning track corresponding to different target crystal direction angles; Each set of fitting coefficients corresponding to different target crystal direction angles is obtained by performing n-order polynomial fitting on the deformation curve; wherein n is an integer, and n≤5; The fitting coefficients corresponding to different target crystal orientation included angles are stored to obtain the pre-stored data set.

4. The wafer topography detection data compensation method of claim 1, wherein, The gravity compensation value is determined based on the first gravity compensation value and the first difference coefficient, the second gravity compensation value and the second difference coefficient, and the measured thickness value of the wafer under test, including: A first product of the first gravity compensation value and the first difference coefficient, and a second product of the second gravity compensation value and the second difference coefficient are determined, and a product sum of the first product and the second product is calculated; A thickness ratio of the thickness value of the target wafer to the measured thickness value of the wafer under test is determined, and the gravity compensation value is determined based on the thickness ratio and the product sum.

5. The wafer topography detection data compensation method of claim 1, wherein, The tilt compensation value of each of the scan trajectories is determined using the first compensation data of the plurality of scan trajectories, including: For each of the scan trajectories, a first equation among the tilt compensation value, the first compensation data and the real value to be solved on the scan trajectory is established to obtain a first equation group; wherein the first equation group includes a plurality of first equations corresponding to the plurality of scan trajectories; A target distance between support points and a distance L from the support points to the center of the wafer under test are obtained, and for each of the scan trajectories, a second equation among the tilt compensation value, the target distance, the distance L and the real value to be solved on the scan trajectory is established to obtain a second equation group; wherein the second equation group includes a plurality of second equations corresponding to the plurality of scan trajectories; the support points are used to support the wafer under test; The tilt compensation value of each of the scan trajectories is determined based on the first equation group and the second equation group and using the least square method.

6. The wafer topography detection data compensation method according to claim 5, wherein The tilt compensation value of each of the scan trajectories is determined based on the first equation group and the second equation group and using the least square method, including: constraint conditions are established for the first equation group and the second equation group based on the real value to be solved on at least two scan trajectories to obtain a joint equation group; the least square method is used for the joint equation group to obtain the tilt compensation value of each of the scan trajectories; For each of the scan trajectories, the first compensation data of the scan trajectory is tilt-compensated by the tilt compensation value to obtain second compensation data of the scan trajectory after tilt compensation, including: for each scan trajectory, the tilt compensation value and the first compensation data of the scan trajectory are substituted into the first equation corresponding to the scan trajectory to obtain the second compensation data of the scan trajectory after tilt compensation.

7. The wafer topography detection data compensation method according to claim 6, wherein The scan trajectories are four, and the current scan trajectory is obtained by scanning the wafer under test based on a crystal orientation included angle corresponding to a previous scan trajectory being rotated by 45 degrees; The second equation group includes four second equations, and the joint equation group includes six equations.

8. A wafer topography measurement apparatus, characterized by, The measurement device includes: A lateral displacement assembly for horizontal movement; A rotating and lifting assembly is used to rotate and / or lift a wafer to be measured, and rotates the wafer by a corresponding angle by rotating and / or lifting the wafer; the rotating and lifting assembly comprises a first rotating and lifting member and a second rotating and lifting member; An up-and-down measuring head assembly comprises two up-and-down distance sensors; A wafer placement table is used to place the wafer to be measured, and is configured as a three-point support structure; the lateral displacement assembly drives the wafer placement table to drive the wafer to be measured to move horizontally between the first rotating and lifting member and the second rotating and lifting member; when the wafer to be measured passes through the two up-and-down distance sensors, the two up-and-down distance sensors collect the height value of the wafer surface to a reference plane in the diameter direction of the wafer to be measured, to obtain the original measurement data s(x) of a plurality of scanning tracks of the wafer to be measured; A control module stores a computer program, and when the computer program is executed, the wafer topography detection data compensation method of any one of claims 1 to 7 is realized.

9. The wafer topography measurement device of claim 8, wherein, The measuring device further comprises a five-axis adjustment mechanism, which is used to perform X / Y-axis translation, Z-axis lifting and pitch yaw adjustment on the two distance sensors; The five-axis adjustment mechanism comprises a Y-axis direction adjustment module for performing Y-axis translation on the two distance sensors, an X-axis direction adjustment module for performing X-axis translation on the two distance sensors, a Z-axis lifting module for performing Z-axis lifting on the two distance sensors, and a pitch yaw adjustment module for performing pitch yaw adjustment on the two distance sensors.

Images