JPH07190741A - Measuring error correction method - Google Patents

Abstract

PURPOSE:To enhance measuring accuracy in a simple constitution by measuring in advance a relation between the position of each stage, and each image at the aforesaid positions, or each error in posture for a measured object, and thereby removing errors made at the time of measurement based on the aforesaid relation. CONSTITUTION:Each coordinate of three marks MA through MC in the Z direction out of marks for the four corners of a two dimensional measuring reference scale 4, is obtained with X and Y axis stages 8 and 6 driven, based on the obtained coordinates, a parallel adjustment for the X Y scanning surface of the scale 4 and a photographing optical system is carried out, and the adjustment of a rotating angle around a Z axis for a Z axis (a Y axis) is also carried out, so that the relative error in posture between the stages 6 and 8 is thereby corrected. Next, the positional coordinate of each mark in the X direction from the mark MA is measured the difference between each aforesaid measured coordinate and the coordinate of each mark which is accurately measured in advance, is stored as measuring errors which depend on the posture of each stage 6 and 8 at the positions of the respective marks. In a practical measurement, a corrected value where each corresponding measuring error is added to each measured value at the measuring positions of the stages 6 and measured value where each measuring error due to the posture of each stage 6 and 8 is removed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an optical measurement method combining image processing and a stage, and relates to a measurement error correction method for enabling highly accurate measurement with a simple apparatus configuration.

[0002]

2. Description of the Related Art Conventionally, in a measuring device for a wide range with high accuracy, a measuring means (for example, an image pickup optical system and an optical method for performing image processing) for obtaining image information of an object to be measured and an object to be measured. Alternatively, a combination of high-precision stages for moving the measuring means is indispensable, and a personal computer or the like is used to further automate the measurement. The high-precision measurement of such a device has adopted a method of measuring the amount of movement in the moving direction of the stage with high accuracy using a laser interferometer or linear encoder.However, the measurement depends on the change in the posture accompanying the movement of the stage. The effect of the error was not corrected.

[0003]

FIG. 2 shows the influence on the measurement error due to the change of the stage posture caused by the movement of the stage. Here, the posture vector of the stage 2 is P, the optical axis posture vector of the imaging optical system 1 is S, and the corresponding vectors before and after the movement are shown with subscripts ( _A ) and ( _B ) respectively.

The stage 2 equipped with the image pickup optical system 1 is
The mark M _A of the DUT 3 is stationary at the position A so as to coincide with the optical axis S _A of the imaging optical system 1, and the posture vector P _A of the stage 2 is parallel to the Y axis. Next, when the stage 2 moves parallel to the two marks M _A and M _B by the same distance as the distance between M _A and M _B and stands still at the position B,
Since the posture vector P _B is inclined by the angle θ with respect to the posture vector P _A , the optical axis vector S _B does not match the mark M _B. The causes of this are the influence of the processing accuracy and assembly accuracy of stage components, the shift of the operating position of the driving force with respect to the center of gravity of the load,
This is due to the non-linearity of stage movement due to mechanical factors such as non-uniformity of frictional force in the sliding portion.

FIG. 3 shows a vector representing the measurement error due to the change in posture at this time. Now, the posture vector P _A before the movement coincides with the Y axis and the optical axis vector S _A coincides with the Z axis, respectively, and the posture vector P A after the movement of the stage 2
_{If B} is tilted by an angle θ with respect to the vector P _A , and as a result, the optical axis vector S _B is also tilted by the same angle with respect to the vector S _A , then the mark M _B has a posture vector P _B and an optical axis vector S _B. It appears to exist in a position determined by the composition of. That is, the measurement error of the mark M _B is equal to the error vector R connecting the coordinate origin O and the tip of the composite vector, and the deviation amounts in the X, Y, and Z directions at this time are given by the following equations.

R _X = R cos β sin α (1) R _Y = R cos β (2) R _Z = R cos β cos α (3) where the angle β is the intersection angle between the error vector R and the XZ plane,
The angle α is the mapping vector R ′ of the error vector R onto the XZ plane.
And the optical axis vector S _A.

On the other hand, even if the object to be measured is moved with the imaging optical system fixed, the posture of the object to be measured changes with the movement of the stage. The difference between the movement of the imaging optical system and the movement of the measured object is the distance from the rotation center of the stage (point P in FIG. 2) to the measured object (corresponding to the length of the error vector R described above). The difference is that the former is longer than the latter. From the viewpoint of measurement error, it is advantageous to move the object to be measured, but on the other hand, it may not be preferable to move the object to be measured. In any case, a measurement error due to the influence of the posture change accompanying the movement of the stage is inevitable.

The present invention has been made in view of the above circumstances, and corresponds to the stop position of the stage and its position by newly using a highly accurate reference scale with the conventional simple structure in which the linear encoder and the stage are combined. Read the posture change of the stage, and store the value in the personal computer,
In the actual measurement, it is an object to provide a measurement error correction method that uses this value to correct the measurement error.

[0009]

In order to achieve the above object, the present invention provides an optical measuring method including an optical system for imaging, a stage and a personal computer for control, the optical system for imaging or the object to be measured. The relationship between the position of the stage and the attitude error of the image or the measured object at that position is measured in advance with respect to the attitude error of the image or the measured object captured by the imaging optical system caused by the movement of the stage on which the object is mounted. According to this relationship, the error in measurement is removed by the control personal computer.

[0010]

According to the present invention, the measurement error of the mark M _B is (1) to (1) if the values of R and the angles α and β are clarified.
Although it can be obtained by calculation from equation (3), it is difficult to obtain these values by actual measurement. Therefore, by measuring this mark position by placing a reference scale having a plurality of marks, the difference between the known mark position previously measured with high accuracy by another method and the mark position measured by this device is compared. Is obtained as a measurement error due to the attitude of the stage at the stop position.

According to this method, the position of the stage stopped for measurement and the posture of the stage at that position are in a one-to-one correspondence.
Therefore, it is important to measure the stage stop position at the smallest possible interval.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings.

FIG. 1 shows an apparatus configuration for measuring the position of a two-dimensional array. Reference numeral 4 is a two-dimensional measurement reference scale, 5 is a goniometer stage capable of adjusting three mutually orthogonal angles, 1
Is an imaging optical system, 6 is a Y-axis stage, 7 is a Y-axis linear encoder, 8 is an X-axis stage, 9 is an X-axis linear encoder, 10 is a Z-axis stage, and 11 is a Z-axis linear encoder.

In order to correct the attitude error of the stage, first, the two-dimensional measurement reference scale 3 and the XY scanning plane of the imaging optical system 1 are adjusted in parallel, and the X-axis (or Y-axis) Z is adjusted.
It is necessary to adjust the rotation angle around the axis (parallel to the imaging optical system 1) to correct the relative attitude error between the two.

The parallelism is adjusted by driving the X-axis stage 8 and the Y-axis stage 6 to form three marks (M _A , M _B , M) out of the four marks formed at the four corners of the two-dimensional measurement reference scale 4.
_c : The position coordinate in the Z direction is calculated for the mark interval of 30 mm). If the mark is out of the depth of focus of the imaging optical system 1, the Z-axis stage is moved so as to capture the mark within the depth of focus, and the amount of movement is read by the Z-axis linear encoder. From the Z coordinates of these three marks,
It is possible to know the error in the parallelism of the two-dimensional measurement reference scale 4 with respect to the XY scanning plane of the imaging optical system 1 and its direction. On the other hand, the rotation angle around the Z-axis is also 3 for the three mark measurements.
It is obtained from the positional deviation in the Y direction (or X direction) caused by the movement of 0 mm in the X direction (or Y direction).

The parallelism error and the rotation angle about the Z axis thus obtained drive the gonio stage 5 to directly adjust the attitude of the two-dimensional measurement reference scale 4. In this measuring device, the depth of focus of the imaging optical system 1 is about 20 μm,
Since the maximum attitude error of the stage is about 2 μm,
In this adjustment stage, the parallelism is 0.7 mrad (20/300
00), the rotation angle is 0.02 mrad (2/30000)
Error will be included.

Next, the mark is measured along the X (+) direction from the mark M _A , and when the mark M _B at the end is reached, one mark is moved in the Y direction, and this time, in the X (−) direction. Measure the mark position again. By repeating this operation in sequence and measuring all mark positions, the measurement error depending on the posture of the stage at the position corresponding to each mark can be obtained.

The two-dimensional measurement reference scale used here is manufactured by the same method as the photomask used in the LSI semiconductor process, and the positions of the marks two-dimensionally arranged at 250 μm intervals are high-precision laser interference. 0.1μ in total
It is measured with an accuracy of m. The method for obtaining a specific measurement error due to the posture of the stage using this reference scale is performed as follows.

The first mark M _A is positioned so that it coincides with the center of the image pickup optical system 1, and is set as a reference point having no measurement error at this position. Then the stage for measuring the second mark M ₂ to 250μm moved in X (+) direction, the position coordinates (X ₂ of the mark M ₂ by the imaging optical system 1,
Y ₂ ) is measured. At this position coordinate (X ₂ , Y ₂ ),
Error 0.0 that could not be removed by adjusting the rotation angle around the Z axis
The effect of 2 mrad is included, but the amount is about 0.0
It is as small as 2 μm (250 μm × 2/30000) and does not pose a problem. The position coordinate (X ₂ , Y ₂ ) of the mark M ₂ is compared with the coordinate (x ₂ , y ₂ ) measured in advance by the high-precision laser interferometer on the two-dimensional measurement reference scale, and the difference Δ between the two. _{x2 (= X 2 -x 2)} , the same as the _{Δ y2 (= Y 2 -y 2} ) is the (1) to (3) measurement error R _x according to the posture of the stage described by the formula, R _y. The positions (X _n , Y _n ) and the measurement errors (Δ _Xn , Δ _Yn ) corresponding to the respective positions are similarly obtained for the third and subsequent marks. This position coordinate (X _n , Y _n ) and measurement error (Δ _Xn , Δ
_Yn ) is set as one set, and these values are stored in the personal computer for all marks.

After that, actual measurement is performed. For example, if the stage obtains measured values (x _j , y _j ) at position coordinates (X _j , Y _j ), a measurement error (Δ _Xj , Δ _Yj ) is added to the measurement value (x _j , y _j ) to obtain a correction value (x _j + Δ _Xj , y _j + Δ _Yj ) which is a measurement value from which the measurement error due to the posture of the stage is removed. If the position where the stage stops for measurement does not match the position coordinate of the mark measured in advance, for example, the actual measurement position coordinate (X _i , Y
_i ) is between the marks M _j [(X _j , Y _j ) (Δ _Xj , Δ _Yj )] and M _K [(X _k , Y _k ) (Δ _Xk , Δ _Yk )], the marks M _j and M. if the distance _k between the position which internally divides m pairs n, the measurement error (Δ _Xi, Δ _Yi) of the measurement position coordinates by interpolation (X _i, Y _i) is given by the following equation.

Δ _Xi = {m / (m + n)} × Δ _Xk + {n /
(M + n)} × Δ _Xj (4) Δ _Yi = {m / (m +
n)} × Δ _Yk + {n / (m + n)} × Δ _Yj (5)
FIG. 4 shows an example of the measurement result of a two-dimensional array (the one-dimensional display shows the measurement value of the Y component with respect to the movement in the X direction) when the influence of the measurement error due to the attitude change of the prototype measurement device is corrected and not corrected. Is. The accuracy when the measurement error due to the attitude of the stage is not corrected is ± 1 μm, but the accuracy of this correction method is ± 0.3 μm, and the effect of improving the measurement accuracy was confirmed.

[0022]

As described above, by using the correction method of the present invention, it is possible to greatly improve the measurement accuracy with a simple structure in which a stage and a linear encoder are combined.
This can be expected to be effective in downsizing the device and reducing manufacturing costs.

[Brief description of drawings]

FIG. 1 is a perspective view showing a configuration of a prototype device according to an embodiment of the present invention for examining a measurement error correction method due to a posture change in measurement of a two-dimensional array.

FIG. 2 is a perspective view showing a conventional measuring apparatus and showing an influence on a measurement error due to a change in posture caused by movement of a stage.

FIG. 3 is an explanatory diagram in which a measurement error due to a conventional posture change is displayed as a vector.

FIG. 4 is a characteristic diagram comparing the measurement results with and without the correction method.

[Explanation of symbols]

1 ... Optical system for imaging, 2 ... Stage, 3 ... Object to be measured, 4 ...
Reference scale for two-dimensional measurement, 5 ... Goniometer stage, 6 ...
Y-axis stage, 7 ... Y-axis linear encoder, 8 ... X-axis stage, 9 ... X-axis linear encoder, 10 ... Z-axis stage, 11 ... Z-axis linear encoder.

Claims (1)

[Claims] 1. An optical measuring method comprising an imaging optical system, a stage and a control personal computer, wherein the imaging optical system or the imaging optical system generated by movement of a stage carrying an object to be measured. For the captured image or the attitude error of the object to be measured, the relationship between the position of the stage and the attitude error of the image or the object to be measured at that position is measured in advance,
Due to this relationship, the measurement error correction method is characterized in that the measurement error is removed by the control PC.