JP4824647B2 - Flatness measurement method - Google Patents

Description

The present invention relates to measuring how the flatness of the surface of the object to be measured such as a semiconductor substrate.

There are several methods for inspecting whether the flatness is not damaged by warping of the measured object such as a semiconductor substrate (wafer), such as an interferometer, a displacement such as a laser displacement meter or a capacitance displacement meter. There is a method of measuring the height (displacement) distribution of the surface of the object to be measured by scanning the meter and obtaining the flatness from the measured value.
A general method for scanning a displacement meter is to perform one-dimensional scanning with a displacement meter in two orthogonal directions on the surface of the object to be measured (in a cross shape in the x and y directions), respectively. The flatness is obtained from the measured values in the direction, and is generally called the cross method. Here, the flatness is evaluated, for example, by obtaining a reference plane such as a least-squares plane from the measured value and by measuring the amount of displacement of each measured value with respect to the reference plane.
At this time, if the number of scans by the displacement meter is increased, more accurate flatness can be measured. However, in order to change the scanning direction, if a rotation mechanism is provided on the support portion of the object to be measured or the support portion of the displacement meter, Surface shake due to looseness of the rotation mechanism is likely to occur, and as a result, a different inclination occurs in the object to be measured for each scan. Also, in the method in which the object to be measured is placed on a predetermined fixed support portion and measured, and the object to be measured is rotated in order to change the scanning direction, the supported point or object on the surface of the object to be measured is also measured. When unevenness is present on the support surface, this causes a different inclination of the object to be measured for each scan. This inclination is also caused by the difference in height between the support point or the support surface that supports the object to be measured. As described above, when a different tilt occurs for each scan, the measured value includes a displacement amount (hereinafter referred to as a correction displacement amount that is required) due to the tilt. The amount of displacement required for correction can be corrected if scanning is performed in two directions. However, if scanning is performed in three or more directions and a different inclination occurs in the object to be measured for each scanning, the reference plane is not determined and the flatness is corrected correctly. The degree cannot be measured.
Therefore, in Patent Document 1, the lower surface of the reference plate provided with a support portion that supports three points on the lower surface of the object to be measured is processed with high accuracy (high flatness), and another support portion is further formed from the lower surface of the reference plate. It is proposed to eliminate surface blurring by rotating the reference plate. Here, when the object to be measured is rotated, the reference plate is rotated together. As a result, the lower surface of the reference plate is processed with high accuracy, so that even when the reference plate is rotated, different inclinations are not generated in the object to be measured.
It is also known that the surface shake can be reduced by a rotating mechanism using an air spindle.
JP-A-7-332962

However, in the method disclosed in Patent Document 1, when the scanning direction is changed, the support portion placed on the reference plate rotates together with the object to be measured, so the surface on the lower surface side of the object to be measured is measured. In this case, there is a problem that the displacement gauge and the support part interfere with each other, and as a result, the lower surface side of the object to be measured cannot be measured. For this reason, the flatness of both the upper and lower surfaces of the object to be measured could not be measured simultaneously. Further, the lower surface of the reference plate is worn with the support portion that supports the reference plate, and it is difficult to maintain a highly accurate finished surface for a long period of time. In addition, the air spindle is generally expensive, and there is a problem that the cost becomes high when it is used.
Accordingly, the present invention has been made in view of the above circumstances, and its object is to measure the flatness of the surface of the object to be measured by performing one-dimensional scanning with a displacement meter from more directions. In this case, the flatness of the lower surface of the object to be measured can be measured without interfering with the support part of the object to be measured, and the flatness can be measured at a low cost without requiring any special effort to maintain the measurement accuracy. It is to provide a measurement how.

In order to achieve the above object, the present invention sequentially supports three support points that support a part of the plurality of supported points on the lower surface of the plate-shaped object to be measured, and each support supports a straight line by a displacement meter. By performing scanning in the direction, the distribution of the surface height of the object to be measured is measured along a plurality of different linear scanning lines on the measurement surface of the object to be measured, and based on the measurement value obtained by the measurement A flatness measurement method for measuring the flatness of the surface of the object to be measured, wherein the displacement meter is configured such that two predetermined support points among the three support points are viewed from a direction substantially perpendicular to the measurement surface. In a state where the measurement object is arranged on a straight line parallel to the scanning direction, the object to be measured is set with a predetermined origin point on the straight line scanned by the displacement meter as viewed from a direction substantially perpendicular to the measurement surface. While rotating sequentially, each of the scanning lines Is viewed along the direction perpendicular to the surface, so as to pass through the positions of two of the object support points being supported by said predetermined two support points either during the scanning of the other of said scanning lines, said are sequentially supported at three or more different orientation of the object to be measured with respect to the three support points, each 3 or more directions comprising said different, and measurement procedure for the measurement by the displacement gauge, for each pre-Symbol the supported point Based on the measurement value obtained by the measurement procedure, the inclination in two directions among the inclinations of the object to be measured in each of the three or more different directions generated by the support at the two predetermined support points is fixed. And a true surface height calculation procedure for calculating a true surface height that is not affected by the inclination of the object to be measured under the above-described conditions.

Incidentally, according to the measurement procedure, all of the scanning lines, the is viewed along the direction perpendicular to the measurement surface of the object to be measured passes through a position of the rotation center.

  Further, when the position of the rotation center is O and the two predetermined support points are A and B, the lengths of the line segments OA and OB are equal, and the scanning direction by the displacement meter and the straight line OA. And the support points A and B are arranged at positions where the angles formed by OB and OB are (180 / N) ° and (180−180 / N) ° (N = an integer of 3 or more) in a predetermined rotation direction, respectively. The scanning lines are N lines rotated by (180 / N) ° about the position O of the center of rotation when viewed from a direction substantially perpendicular to the measurement surface of the object to be measured. It is also possible to do.

According to the present invention, it is possible to perform one-dimensional scanning by the displacement gauge from more directions determine an accurate flatness of the workpiece surface. (That is, as will be described later as processing in steps S6 and S7, true surface height data can be calculated from the measured values for N (N ≧ 3) scanning lines, and the true surface height can be calculated. (Accurate flatness can be obtained from the data.) In addition, it is possible to measure the lower surface of the object to be measured. Furthermore, in order to maintain the measurement accuracy, no special labor such as maintaining a high-precision machined surface is required, and flatness can be measured at a low cost because no expensive device is required.

Hereinafter, embodiments and examples of the present invention will be described with reference to the accompanying drawings so that the present invention can be understood. It should be noted that the following embodiments and examples are examples embodying the present invention, and do not limit the technical scope of the present invention.
Here, FIG. 1 is a three-sided view showing a schematic configuration of a flatness measuring apparatus U according to an embodiment of the present invention, and FIG. 2 measures an object to be measured in the flatness measuring apparatus U according to an embodiment of the present invention. FIG. 3 schematically shows the surface height of the object to be measured in the flatness measuring apparatus U according to the embodiment of the present invention, and FIG. 4 shows the flatness according to the embodiment of the present invention. FIG. 5 is a diagram showing a support point of the object to be measured in the measuring device U1, FIG. 5 is a diagram showing a scanning line for measuring the object to be measured in the flatness measuring device U1 according to the embodiment of the present invention, and FIG. 6 is an embodiment of the present invention. FIG. 7 is a diagram showing a support point of an object to be measured in the flatness measuring device U2 according to FIG. 7, and FIG. 7 is a diagram showing a scanning line for measuring the object to be measured in the flatness measuring device U2 according to the embodiment of the present invention.

First, a schematic configuration of the flatness measuring apparatus U according to the embodiment of the present invention will be described with reference to FIG.
The flatness measuring device U measures the flatness of a circular plate-shaped semiconductor substrate (wafer) which is an example of a measurement object. Of course, the wafer is not limited to a circular plate.
The wafer 1 is supported substantially horizontally at the tips (support points A to C) of the three support portions 2a to 2c, and in this state, the capacitance type displacement sensor 3 (hereinafter simply referred to as a sensor) is set to be substantially horizontal. Scanning in the linear direction (negative direction of the Y-axis), the surface height (displacement amount in the Z-axis direction not shown) in the substantially vertical direction of the upper and lower surfaces of the wafer 1 is simultaneously measured. Hereinafter, for the sake of convenience, the measured value on the lower surface side of the wafer 1 is simply referred to as the measured value. If the shape (surface height distribution) of the lower surface of the wafer 1 can be measured by the measured value on the lower surface side, the thickness of the wafer 1 can also be measured by the difference between the measured values on the upper surface side and the lower surface side. If the thickness of the wafer 1 is known (eg, it is guaranteed that the wafer 1 is manufactured with a predetermined accuracy), the shape of the upper surface can be measured by adding the thickness to the measured value on the lower surface side.
The support points A and B (the two support points) at which the two support portions 2a and 2b out of the three support portions 2a to 2c come into contact with the wafer 1 are in the Z-axis direction (a direction substantially perpendicular to the surface of the wafer 1). ) (A plan view of FIG. 1), they are on a straight line parallel to the scanning direction of the sensor 3 (the negative direction of the Y-axis) and 45 ° and 135 ° (clockwise from the positive direction of the Y-axis, respectively) 180 [deg.]-45 [deg.].
The position of the remaining support portion 2c is not particularly limited as long as it can stably support the wafer 1 together with the support portions 2a and 2b. Here, it is arranged at an appropriate position on the opposite side of the intermediate position between the support portions 2a and 2b and the Y-axis.
Although not shown in FIG. 1, the flatness measuring device U lifts the wafer 1 supported by the support portions 2a to 2c and, when viewed from above (the positive direction of the Z axis), has a predetermined value on the Y axis. There is an elevating / rotating device for rotating the wafer 1 by a predetermined angle around the origin O (corresponding to the rotation center point) and then lowering the wafer 1 and supporting it on the support portions 2a to 2c.
In the flatness measuring device U, the wafer 1 placed on the support portions 2a to 2c is rotated by 45 ° clockwise as viewed from above by the elevating / rotating device in four directions (180 ° / 45 °). ) Direction).
The measured values on the upper surface side and the lower surface side are input to a computer (not shown), and various calculations for obtaining the flatness of the wafer 1 are performed by the computer. Further, the computer controls these by outputting control signals to the sensor 3 and the lifting / rotating device.
Thus, since the relative positional relationship between the support portions 2a to 2c and the sensor 3 is made constant and only the wafer 1 is rotated to change the scanning direction, the sensor 3 and the support portions 2a to 2c interfere with each other. The lower surface side of the wafer 1 can be measured without any problems. In addition, since there is no high-precision finished surface with wear, there is no need to take measures against wear. Furthermore, it is a general one-dimensional scanning type measuring apparatus that does not use an expensive apparatus such as an air spindle, and is low in cost.
That is, the present invention devises a measurement position (scan line position), a support position (support points A and B), and a measurement value processing method described below using a general primary scanning type measurement apparatus. By doing so, the exact flatness of the wafer 1 is obtained.

Next, the procedure for measuring the flatness of the surface of the wafer 1 in the flatness measuring apparatus U will be described with reference to FIG. Hereinafter, S1, S2, and so on represent processing procedure (step) numbers. Before the processing of FIG. 8 is started, the wafer 1 is placed on the supporting portions 2a to 2c in a predetermined direction (for example, a crystal direction (notch direction or orientation flat direction) of the wafer 1) in advance by manual operation or a predetermined apparatus. In the positive direction of the Y-axis, etc.) on the support portions 2a to 2c.
First, the sensor 3 is scanned, the surface height of the wafer 1 is measured (S1), the measured value is taken in by the computer, and the error correction (S2) of the measured value is performed.
The error correction is a process of removing errors caused by gravity deflection of the wafer 1, scanning errors of the sensor 3, errors caused by a change in capacitance between the wafer 1 and the sensor 3 due to the rotational movement of the wafer 1, and the like from the measured values. It is. In the error correction, a calibration reference wafer having a uniform thickness is measured a plurality of times in advance, an average value of the measurement values is stored in the computer as calibration data, and the measurement value obtained in S1 using the calibration data. Correct.
Next, the measured value subjected to the error correction is stored by the computer (S3).
Next, it is determined (S4) how many times the measurement performed in S1 by the computer is for the wafer 1, and if it is determined that the measurement is less than N times (N = 4 in this case). The wafer 1 is rotated by (180 / N) ° (ie, 45 °) by the lifting / rotating device (S8), and the process returns to S1 to repeat the measurement.
If it is determined in S4 that the measurement is the Nth (N = 4) measurement, the computer performs offset correction on the measurement value stored in S3.
The offset correction means that all the measurement values at the position of the origin O (rotation center of the wafer 1) measured in common in all N times (N = 4) measurements are equal. It is to correct the offset amount. Specifically, for each scan, a value obtained by subtracting the measurement value at the origin O from each measurement value is set as a measurement value after correction for the offset (hereinafter referred to as a relative measurement value). As a result, all the measurement values at the position of the origin O become 0 (zero), and the relative measurement value indicates a relative height with respect to the measurement value of the origin O. In this way, the correction to the relative measurement value is performed in the flatness measurement because the absolute value of the displacement (surface height) in the Z-axis direction of the surface of the wafer 1 is not necessary and is a reference. This is because it is only necessary to grasp the relative height distribution of the surface of the wafer 1 with respect to the plane.
Next, by substituting (applying) the relative measurement value obtained in S5 into a predetermined equation to be described later, a difference between the true surface height of the wafer 1 surface and the relative measurement value (necessary correction displacement amount) is obtained. Obtain (S6).
The relative measurement value is generated by unevenness present at a point (supported point) on the surface of the wafer 1 supported by the support points A and B and a difference in height between the support points A and B. This is a value obtained by stacking the displacement amount due to the inclination of 1 (the correction displacement amount required). For this reason, the relative measurement value cannot be a value of the shape (surface height) of the wafer 1. The true surface height means a distance (height) from the plane assuming a plane having a certain positional relationship with the wafer 1. That is, it represents a height that is not affected by the amount of displacement caused by the tilt of the wafer 1.
Finally, a reference plane such as a least-squares plane is obtained from the true surface height data obtained in S6, and the magnitude of the difference between the true surface height and the height of the reference plane (for example, standard deviation) The flatness of the surface of the wafer 1 is evaluated based on the maximum difference (S7).

Next, equations used in S6 of FIG. 8 will be described.
First, symbols used below will be described with reference to FIG.
FIG. 2 is a schematic diagram when the wafer 1 as the object to be measured is viewed from above (in the positive direction of the Z axis). The thick line indicates a line (scanning line) scanned by the sensor 3, and the thick line arrow indicates the sensor. 3, points h1 to h8 represent supported points which are points on the surface of the wafer 1 supported by the supporting points A and B in four scans of the sensor 3. The supported points h1 to h8 are actually on the lower surface side of the wafer 1 at the position shown in FIG.
The scan line Dir0 in the first scan passes through supported points h1 and h5, and the scan line Dir45 in the second scan passes through supported points h2 and h6, and the scan in the third scan. Dir 90 passes through supported points h3 and h7, and the scanning line Dir135 in the fourth scan passes through supported points h4 and h8. The angle θ formed by the adjacent scanning lines is 45 °. Further, the distance from the origin O to each of the supported points h1 to h8 is L (therefore, for example, 2L between h1 and h5), and between h1 and h3, between h2 and h4, and between h6 and h8. The distance, that is, the distance between the support points A to B in FIG.

Next, the relationship between the true surface height and the relative measurement value is formulated.
For example, in the first scanning, the wafer 1 is inclined due to the difference in height between the supporting points A and B and the unevenness of the supported points h2 and h4. This inclination differs for each scan.
Here, the true surface height at the supported point hi (i = 1 to 8) excluding the influence of the inclination is wi, the relative measurement value at the supported point hi is si, and the difference therebetween. The amount of displacement required for correction is ki (= si-wi).
FIG. 3 is a graph showing the true surface height (broken line Wg) and the relative measurement value (solid line Sg) in the first scan (scan line Dir0). The vertical axis (Z axis) of the graph represents the surface height of the wafer 1, and the horizontal axis (Y axis) represents the scanning direction by the sensor 3. As described above, since the distances from the origin O to the supported points h1 and h5 are equal, if the true surface height Wg at the origin O is matched with the relative measurement value Sg, the supported points If the correction required displacement amount at h1 is + k1, the correction displacement amount at the supported point h5 is −k1.
When this relationship is expressed with respect to the supported points h1 to h8, the following equation 1 is obtained.

ここで，ｒ＝Ｌ／ｄ＝１／ｃｏｓ（４５°）＝√２である。 Next, the correction required displacement amounts k1 to k4 are formulated.
Hereinafter, the first scanning (corresponding to the scanning line Dir0) will be described.
First, if the difference in height between the support points A and B is 2Δ (the height of A−the height of B), the distance in the Y-axis direction between the points A and B is 2d. The inclination in the Y-axis direction caused by the height difference is (2Δ / 2d).
Here, since the line passing through the supporting points A and B (corresponding to the points h4 and h2 in FIG. 2) and the scanning line Dir0 are parallel when viewed from the Z-axis direction, the scanning line Dir0 The inclination of the direction is also (2Δ / 2d). Therefore, the displacement amount caused by the inclination of the wafer 1 due to the difference in height between the points A and B at the point h1 is that the position of the point h1 is the L position in the negative direction of the Y axis from the origin O. , {− (2Δ / 2d) · L}.
Further, the inclination of the wafer 1 in the direction of the scanning line Dir0 due to the unevenness on the surface of the wafer 1 is the inclination {− (w4− w2) / 2d}. (When the height of the surface of the wafer 1 in the Z-axis direction is high, that is, when it is in a concave state when viewed from the lower surface side of the wafer 1, the sign of the inclination is reversed because the wafer 1 sinks in the negative direction of the Z-axis. Therefore, at the point h1, the amount of displacement caused by the inclination of the wafer 1 due to the irregularities on the surface of the wafer 1 is {(w4-w2) / 2d · L}. The required displacement amount k1 at the point h1 is a value obtained by adding these two displacement amounts.
In the above manner, the required correction displacements k1 to k4 in the respective scans (the scanning lines Dir0 to Dir135) are expressed by the following formula 2.

Here, r = L / d = 1 / cos (45 °) = √2.

式３により，未知数ｋ２，ｋ４，Δが３つで方程式が４つとなり，適当な３つの方程式を選択すれば解を求めることができる。しかし，精度を上げるためには，４つの方程式全てを計算結果に反映することが望ましい。そのため，当該平坦度測定装置Ｕでは，３変数１次方程式を求める一般的な最小二乗法により，未知数ｋ２，ｋ４，Δを求める。
このようにして求めた前記要補正変位量ｋ２及びｋ４により，前記走査線Ｄｉｒ４５，Ｄｉｒ１３５の各走査時のＹ軸方向のウェハ１の前記真の傾きが各々，（−ｋ２／Ｌ），（−ｋ４／Ｌ）と求まるので，これにより，前記走査線Ｄｉｒ４５，Ｄｉｒ１３５の各走査時の前記相対測定値を補正すれば，ウェハ１下面の４方向の真の形状（表面高さ）を表すデータが求まる。このようにして求めた，真の形状を表すデータを用いて，例えばそのうちの特定の３点のデータで決定される平面，或いは最小二乗面等（基準平面）からの変位量を求め，該変位量によりウェハ１下面（測定平面）の平坦度（ソリ）が評価できる。 Next, if the correction displacements k1 to k4 required are obtained, the true surface height can be obtained. However, even if Equation 1 is substituted into Equation 2, there are five unknowns (k1 to k4, Δ) and the equation is There are four, and no solution is found.
In order to solve this, it is conceivable to fix the true tilt of the wafer 1 (the true surface height = the tilt of the wafer 1 when the relative measurement value is obtained) and obtain the solution. When the warpage of the wafer 1 is evaluated, the evaluation is based on the amount of displacement from a plane (a plane determined by specific three points of the measurement values, or a least-squares surface) specified from the measurement value of the wafer 1 itself. In other words, since it is evaluated by relative evaluation of the entire measurement value, there is no problem in fixing the true inclination of the wafer 1.
As the method, for example, by setting w1 = w5 or k1 = a (constant), the inclination in one direction among the true inclinations of the wafer 1 is fixed, and the unknowns are set to four and the equations are set to four. It is conceivable to obtain the unknowns k2, k4, Δ. Further, for example, w1 = w5 and w3 = w7 or the like, or k1 = k3 = a (constant) or the like, the true inclination of the wafer 1 is fixed in two orthogonal directions (that is, the true of the wafer 1 is It is also conceivable that the remaining unknowns are obtained by a general least square method for obtaining a three-variable linear equation with the entire slope fixed.
Further, for example, the difference 2Δ between the heights of the support points A and B may be measured in advance, and thereby the unknowns k1 to k4 may be obtained.
Here, k1 = k3 = 0, and k2 and k4 are determined so as to match the relative measurement values without correcting the relative measurement values in the measurement on the scanning lines Dir0 and Dir90. That is, by substituting Equation 1 into Equation 2 and setting k1 = k3 = 0, the following Equation 3 is obtained.

According to Equation 3, there are three unknowns k2, k4, Δ and four equations, and a solution can be obtained by selecting appropriate three equations. However, in order to improve accuracy, it is desirable to reflect all four equations in the calculation results. Therefore, in the flatness measuring apparatus U, the unknowns k2, k4, and Δ are obtained by a general least square method for obtaining a three-variable linear equation.
The true inclinations of the wafer 1 in the Y-axis direction during the scanning of the scanning lines Dir45 and Dir135 are respectively (−k2 / L), (−) based on the correction displacement amounts k2 and k4 thus determined. k4 / L). Accordingly, if the relative measurement value at the time of each scanning of the scanning lines Dir45 and Dir135 is corrected, data representing the true shape (surface height) in the four directions of the lower surface of the wafer 1 is obtained. I want. Using the data representing the true shape obtained in this way, for example, the amount of displacement from a plane determined by the data of three specific points, the least square surface, etc. (reference plane) is obtained, and the displacement The flatness (warp) of the lower surface (measurement plane) of the wafer 1 can be evaluated by the amount.

また，前述した式２に相当する式は，次式２ａとなる。

ｋ１＝ｋ２＝０として，式１ａ及び式２ａを整理すると，前述した式３に相当する式として次の式３ａとなる。

式３ａは未知数ｋ３，Δが２つ，方程式が３つとなり，２変数１次方程式を求める一般的な最小二乗法により解が求まる。
同様に，図６及び図７に示す応用例では，前述した式１〜式３に相当する式が，各々次の式１ｂ，式２ｂ，式３ｂとなる。これらの式も同様に解くことができる。

また，前記平坦度測定装置Ｕでは，センサ３を走査させているが，センサ３とウェハ１の相対移動であればよく，センサ３を固定し，前記支持部２ａ〜２ｃを直線的に移動させるよう構成してもよい。
また，センサ３は，静電容量式変位センサとしたが，例えば，三角測量に基づくレーザ変位計や渦流センサ等，他の変位センサであってももちろん適用可能である。 In the flatness measuring apparatus U, the wafer 1 is rotated 45 degrees at a time and measured in four directions. As an application example thereof, as shown in FIG. 4, the reference support points A and B are moved from the positive direction of the Y axis. The scanning lines may be arranged at positions of 60 ° (180 ° / N, N = 3) and 120 ° in the clockwise direction, and three scanning lines may be provided at intervals of 60 ° as shown in FIG. In addition, as shown in FIG. 6, the reference support points A and B are arranged at 22.5 ° (180 ° / N, N = 8) and 157.5 ° counterclockwise from the positive direction of the Y axis. Then, as shown in FIG. 7, the number of scanning lines may be eight at intervals of 22.5 °.
In the examples of FIGS. 4 and 5, the equation corresponding to the above-described equation 1 is the following equation 1a.

Further, the expression corresponding to the above-described expression 2 is the following expression 2a.

When k1 = k2 = 0 and formulas 1a and 2a are rearranged, the following formula 3a is obtained as a formula corresponding to formula 3 described above.

Equation 3a has two unknowns k3, Δ and three equations, and a solution can be obtained by a general least square method for obtaining a two-variable linear equation.
Similarly, in the application examples shown in FIGS. 6 and 7, the expressions corresponding to the expressions 1 to 3 described above become the following expressions 1b, 2b, and 3b, respectively. These equations can be solved similarly.

Further, in the flatness measuring apparatus U, the sensor 3 is scanned, but it is sufficient that the sensor 3 and the wafer 1 are moved relative to each other, the sensor 3 is fixed, and the support portions 2a to 2c are moved linearly. You may comprise.
The sensor 3 is an electrostatic capacitance type displacement sensor, but other displacement sensors such as a laser displacement meter and a eddy current sensor based on triangulation are also applicable.

  The present invention can be used for measuring the flatness of the surface of an object to be measured such as a semiconductor substrate.

3 is a three-view diagram illustrating a schematic configuration of a flatness measuring apparatus U according to an embodiment of the present invention. FIG. The figure showing the scanning line which measures the to-be-measured object in the flatness measuring apparatus U which concerns on embodiment of this invention. The figure which represents typically the surface height of the to-be-measured object in the flatness measuring apparatus U which concerns on embodiment of this invention. The figure showing the support point of the to-be-measured object in the flatness measuring apparatus U1 which concerns on the Example of this invention. The figure showing the scanning line which measures the to-be-measured object in the flatness measuring apparatus U1 which concerns on the Example of this invention. The figure showing the support point of the to-be-measured object in the flatness measuring apparatus U2 which concerns on the Example of this invention. The figure showing the scanning line which measures the to-be-measured object in the flatness measuring apparatus U2 which concerns on the Example of this invention. The flowchart showing the measurement procedure of the flatness in the flatness measuring apparatus U which concerns on embodiment of this invention.

Explanation of symbols

1. Semiconductor substrate (wafer)
2a to 2c ... support portions A to C ... support points 3 ... capacitive displacement sensors h1 to h8 ... supported points Dir0, Dir45, ... scanning line Sg ... relative measurement value Wg ... true surface height

Claims (2)

By sequentially supporting three support points that support a part of the plurality of supported points on the lower surface of the plate-shaped object to be measured, and performing scanning in a linear direction with a displacement meter for each of the supported points, the measured object The surface height distribution of the object to be measured is measured along a plurality of different linear scanning lines on the measurement surface of the object, and the flatness of the surface of the object to be measured is determined based on the measurement value obtained by the measurement. A flatness measuring method for measuring,
Two predetermined support points among the three support points are arranged on a straight line parallel to the scanning direction of the displacement meter when viewed from a direction substantially perpendicular to the measurement surface. Each of the scanning lines is viewed from a direction substantially perpendicular to the measurement surface while the object to be measured is sequentially rotated about a position of a predetermined origin on a straight line scanned by the displacement meter as viewed from the vertical direction. Te, so as to pass through the positions of two of the object support points being supported by said predetermined two support points either during the scanning of the other of said scanning lines, supporting the object to be measured of the three A measurement procedure for sequentially supporting the point in three or more different directions and performing the measurement by the displacement meter in each of the three or more different directions;
For each of the supported points, out of the inclinations of the measured object in each of the three or more different directions caused by the support at the two predetermined support points based on the measurement values obtained by the measurement procedure A true surface height calculation procedure for calculating a true surface height that is not affected by the inclination of the object to be measured under a condition in which the inclination in two directions is fixed;
A flatness measuring method characterized by comprising: When the position of the center of rotation is O and the two predetermined support points are A and B when viewed from a direction substantially perpendicular to the measurement surface of the object to be measured, the lengths of the line segments OA and OB are And the angle formed between the scanning direction by the displacement meter and the straight lines OA and OB are (180 / N) ° and (180−180 / N) ° (N = an integer greater than or equal to 3) in predetermined rotation directions, respectively. ) To place the support points A and B at the position
2. Each of the scanning lines is N lines rotated by (180 / N) ° about a position O of the center of rotation when viewed from a direction substantially perpendicular to a measurement surface of the object to be measured. The flatness measuring method described in 1.

Images