JPH11325882A - Method of processing wafer flatness data - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable estimating flatness at wafer chucking, etc., in a wafer treating process by sectioning a wafer into cells of adequate size and mutually relating the flatness to the warpage quantity every sectioned cell. SOLUTION: The wafer flatness data process is made by sectioning e.g. an Si wafer into a plurality of rectangular cells of adequate size, for obtaining flatness data about each of them from the heights of points in the vacuum chucking using a flatness measuring unit, and warpage data from the warpage heights at the points using a warp measuring unit. These data are plotted with the warpage quantity on the abscissa and flatness on the ordinate to obtain the relation between the warp quantity and flatness. From relation of both thus obtd. e.g. the wafer surface irregularities on a chuck board in the exposure process can be estimated, based on the measured result of the simpler measurable warpage quantity, without measuring the actual flatness.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data processing method for wafer flatness such as flatness and warpage of a silicon wafer or the like. Here, the flatness of the wafer indicates a state of the surface of the wafer when the wafer is sucked by the chucking machine, and the warpage of the wafer is a state in which the wafer is not sucked (for example, the surface is flat and horizontal). (In a natural state simply placed on a base plate held on a base plate), and in each case, the surface irregularities are expressed based on certain criteria as shown in Table 1 below. Evaluate by the value of height.

[0002]

[Table 1]

[0003]

2. Description of the Related Art A silicon wafer is manufactured from a rod-shaped silicon single crystal through manufacturing processes such as cutting, lapping, etching, polishing, and the like. The influence of these processes on the wafer (for example, during the manufacturing process). Conventionally, the evaluation of the occurrence of stress) has been performed by simply measuring the flatness of the wafer and comparing the measurement result with the measurement result obtained by another measuring device.

[0004]

However, for example, in the exposure step, which is one of the manufacturing processes, since the vacuum chuck plate in the flatness measuring device and the vacuum chuck plate in the exposure device are different from each other, the flatness in the exposure device is low. It was unclear what they were and did not always yield meaningful results. For example, when there is no warpage in the wafer, it is obvious that the flatness of the surface of the wafer depends to some extent on the surface accuracy of the chuck board, and these studies are meaningless. That is, when the wafer has a certain degree of warpage and there is a possibility that a sufficient flatness may not be obtained due to the suction, it is meaningful to study them.

Further, since the flatness measurement is performed in a state in which the wafer is forcibly sucked, a step for preventing the presence of foreign matter on the back surface of the wafer is required. It was difficult to do with all wafers.

SUMMARY OF THE INVENTION The present invention has been made in consideration of the above-mentioned problems, and an object of the present invention is to absorb a wafer in a wafer processing process by organically combining data such as the size and flatness of a wafer. An object of the present invention is to provide a wafer flatness data processing method capable of predicting the flatness at the time.

[0007]

In order to achieve the above object, a wafer flatness data processing method according to the present invention divides a wafer into cells of an appropriate size, and for each of the divided cells, the flatness and the amount of warpage. Are plotted on the vertical and horizontal axes, respectively, so that the flatness and the amount of warpage are associated with each other.

According to the above-mentioned wafer flatness data processing method, it is possible to predict the flatness when a wafer is sucked in a wafer processing process based on the amount of warpage, and the flatness which is difficult to measure during the manufacturing process. Can be predicted without performing an actual flatness measurement. Then, the influence of the flatness during the process can be predicted. Further, the influence of vacuum suction and the accuracy of vacuum suction when transferring a circuit pattern onto a wafer can be easily evaluated.

[0009]

Embodiments of the present invention will be described with reference to the drawings. 1 to 7 show one embodiment of the present invention. First, FIG. 1 schematically shows a configuration for carrying out a wafer flatness data processing method of the present invention. In FIG. 1, 1 is a flatness measuring device, 2 is a warpage measuring device, and 3 is both Measuring device 1,
A data processing unit for appropriately processing the flatness data and the warpage data input from the computer 2, for example, a personal computer.

The flatness measuring device 1 is configured as shown in FIG. That is, reference numeral 4 denotes a chuck board on which the wafer 5 to be measured is placed and which is vacuum-sucked. It is configured to move in a two-dimensional direction with respect to a direction (a direction perpendicular to the paper surface in FIG. 2). Above the chuck 4, a laser light source 7 for emitting a laser beam 6 for measurement, a lens 8, a half mirror 9, a collimator lens 1
0, the reference plate 11 is provided from top to bottom in this order,
On the optical path side divided by the half mirror 9, a CCD camera 12 as a light detection unit is provided.

In the flatness measuring device 1 having the above configuration,
The flatness of the wafer 5 can be obtained by analyzing interference fringes generated in the CCD camera 12 when the laser light 6 is emitted from the laser light source 7.

The warpage measuring device 2 is configured as shown in FIGS. That is, in FIG. 3, reference numeral 13 denotes a horizontal holding base for simply mounting the wafer 5, and a laser light source 15 for emitting a laser beam 14, a lens 16 for converting the laser beam 14 into parallel light, As shown in FIG. 4A, a mask 19 on which a pattern 18 in which a plurality of holes 17 are arranged in a rectangle is formed, a half mirror 20 is provided in this order from top to bottom, and the optical path side divided by the half mirror 20 is provided. In addition, a CCD camera 22 as a light detection unit is provided via a condenser lens 21.

In the warpage measuring device 2 having the above-described structure, the laser light 14 emitted from the laser light source 15 passes through the lens 16 and becomes parallel light.
Pass through the half mirror 20 and vertically enter the surface of the wafer 5, and a part of the light reflected here is transmitted through the half mirror 20, but other light is reflected by the half mirror 20. Then, the light passes through the condenser lens 21 and enters the CCD camera 22.

When the wafer 5 is flat (flat) as shown by a solid line in FIG.
The image obtained in 2 is the mask 1 shown in FIG.
9 is obtained, but the same pattern 18 is obtained.
However, as shown by an imaginary line in FIG. 3, when a part thereof is not flat due to bending upward or the like, an image obtained by the CCD camera 22 is, as shown in FIG.
It becomes distorted. Then, by integrating the amount of distortion in the distorted image, the degree of warpage of the wafer 5 can be obtained quantitatively. In FIG. 4B, reference numeral 17a indicates a distorted hole.

Although not shown, the flatness measuring device 1 and the warpage measuring device 2 are each provided with an arithmetic control unit having a function of controlling the entire device and storing measurement data. These arithmetic and control units are configured to be able to exchange signals with the personal computer 3 as a data processing unit.

Next, a method of processing wafer flatness data using the apparatus having the above configuration will be described with reference to FIGS.
This will be described with reference to FIG. FIG. 5 shows an example of the wafer 5 to be measured. The wafer 5 is divided into a plurality of rectangular (square in the illustrated example) regions 23 (referred to as cells) having an appropriate size.

First, when measuring the flatness of the wafer 5 using the flatness measuring device 1, the wafer 5 is placed on the chuck board 4. At this time, the position of each cell 23 on the wafer 5 is stored in the data processing unit 3 based on the orientation flat 24 or the notch 25 of the wafer 5.

Then, the wafer 5 on the chuck 4 is vacuum-sucked, the height of the surface at each point on the wafer 5 is measured, and the height of each point (flatness data) based on the reference shown in Table 1. Is calculated. This flatness data is sent to the data processing unit 3 and stored.

Next, the amount of warpage of the wafer 5 is measured using the warp measuring device 2. At this time, the wafer 5 is held on the holding base 13 so as to match the position data stored in the flatness measurement. It is set on the top, the height of the warp at each point where the flatness measurement was performed is measured, and the height (warp data) of each point is obtained by calculation based on the criterion exemplified in Table 1. This warpage data is sent to the data processing unit 3,
It is memorized.

FIG. 6 is a diagram for explaining the case where the least square plane reference of the "surface reference" is used as the reference. In this example, the GFLR in one cell 23A is shown.
(Maximum value−Minimum value) is shown, and in this figure, reference numeral 26 denotes a least square plane in the cell 23, P ₁ , P
Reference numeral ₂ denotes a point indicating a plus-side maximum value and a point indicating a minus-side minimum value with respect to the least square plane 26, respectively.

Then, a local slope (LS) in each of the cells 23 is obtained. In general, the flatness has a smaller surface distortion than the warpage. However, when the flatness is expressed by the LS, the entire warp (bend) is less than the individual unevenness.
It is reflected in the value of S. Since the width of the cell 23 is constant,
LS reflects the magnitude of the curvature of the warp in each cell 23.

FIG. 7 shows the relationship between the amount of warpage and the flatness of the warped wafer 5, in which the horizontal axis indicates the amount of warpage and the vertical axis indicates the flatness. Then, the warp increases toward the right on the horizontal axis, and the flatness deteriorates toward the top on the vertical axis. FIG. 7 is obtained by plotting the flatness data and the warpage data at the same point obtained in the flatness measurement and the warpage measurement on the ordinate and the abscissa, respectively.

According to FIG. 7, when the wafer 5 is attracted to the chuck 4 from the curvature and height of the wafer 5,
It is possible to estimate how the flatness of the wafer 5 is affected.

That is, based on the measurement result of the warpage measuring device 2 which can perform the measurement more easily than the flatness measuring device 1, for example, the unevenness of the surface of the wafer 5 on the chuck board in the exposure step is measured. The state can be inferred.

The wafer 5 is chucked in various manufacturing steps. By using the above-mentioned data, the flatness when the wafer 5 is sucked can be predicted, and the flatness during those processes can be predicted. It is also possible to predict the effect of the degree. For example, the effect of vacuum suction and the accuracy of vacuum suction when transferring a circuit pattern to the wafer 5 can be easily evaluated.

The present invention is not limited to the above-described embodiment. For example, various criteria listed in Table 1 are used as appropriate when measuring the flatness and the amount of warpage of the wafer 5. Alternatively, other criteria can be used.

As a device for measuring the flatness and the amount of warpage, it goes without saying that devices other than the above-described devices 1 and 2 may be used.

[0028]

According to the wafer flatness data processing method of the present invention, since the flatness of the wafer and the amount of warpage are organically related, the flatness when the wafer is sucked in the wafer processing process is predicted. This eliminates the need to actually measure the flatness of the wafer, and can eliminate or greatly reduce the complicated flatness measurement step in the wafer manufacturing process.

By effectively utilizing the above-described wafer flatness data processing method, it is possible to mass-produce high quality wafers at low cost.

[Brief description of the drawings]

FIG. 1 is a diagram schematically showing a configuration for implementing a wafer flatness data processing method of the present invention.

FIG. 2 is a diagram schematically showing an example of a flatness measuring device.

FIG. 3 is a diagram schematically illustrating an example of a warpage measuring device.

FIG. 4 is a diagram for explaining the operation of the warpage measuring device.

FIG. 5 is a plan view showing an example of a wafer.

FIG. 6 is a diagram for explaining flatness measurement.

FIG. 7 is a diagram illustrating an example of a relationship between a warpage amount and flatness.

[Explanation of symbols]

 5 ... wafer, 23 ... cell.

Claims (1)

[Claims] 1. dividing a wafer into cells of an appropriate size;
A wafer flatness data processing method, wherein the flatness and the warpage are associated with each other by setting the flatness and the amount of warpage on the vertical axis and the horizontal axis for each of the divided cells.