JPH0451969B2 - - Google Patents

Description

Detailed Description of the Invention [Field of Application of the Invention] The present invention relates to an automatic focusing method used in a device for positioning a target object, such as a wafer in a semiconductor exposure device, on a predetermined reference plane, and the like. It is related to the device.

The above target object may be a workpiece, an object to be inspected, a processing tool, or a measuring tool, and can be applied to various objects.

[Prior art]

For example, in semiconductor exposure equipment, 2 to 3 μm
It is necessary to transfer a fine mask pattern onto a wafer with high resolution.
It is necessary to position within the plane of m.

FIG. 1 shows an example of the semiconductor exposure apparatus described above. The chuck 2 that adsorbed the wafer 1 is at stage 3.
The above-mentioned three members have a structure that allows them to reciprocate between the N position and the M position shown in the figure as indicated by the arrow C-d. This figure shows three members (1,
2, 3) and the three members at the M position are both drawn with solid lines, and the number of each of these three members installed per unit is one.

In FIG. 1, P is the exposure optical axis. The exposure light enters the mask 8 along the optical axis P, passes through the projection optical system 7, and exposes the wafer 1 which is suctioned and fixed to the chuck 2 mounted on the stage 3 at the M position by the vacuum V. Transfer mask pattern to wafer 1
Transfer to. Here, the wafer 1 needs to be positioned.

For this purpose, the wafer 1 is previously positioned in a plane at the N position. That is, the wafer 1 is suctioned and fixed to a freely tiltable chuck 2 using a vacuum V, and the chuck 2 is mounted on a stage 3. The plane positioning device 4 is attached to the observation microscope 5, and the positioning pins 6 are adjusted so as to form a plane within the depth of focus of the projection optical system 7. Furthermore, the stage 3 has a structure that allows it to move with high precision in the directions of arrows a, b, c, and d. First, move the stage 3 in the direction of arrow a (upward) with the spherical seat 2a of the chuck 2 moving freely.
The surface of the wafer 1 is pressed against the positioning pins 6 to align the wafer 1 and the mask 8 in parallel. Next, the spherical seat 2a of the chuck 2 is fixed to the stage 3 by a vacuum V, and the stage 3 is moved by h in the direction of arrow b (downward). As a result, the wafer 1 is positioned within the depth of focus of the image of the mask 8 located at a position h below the positioning pin 6. In this state, alignment in the horizontal plane is performed, and the relative positioning of the wafer 1 with respect to the mask pattern is completed. Next, move the stage 3 horizontally in the d direction,
Wafer 1 is exposed at position M.

In order to transfer a fine mask pattern to wafer 1 with high resolution, the exposure surface of wafer 1 is kept within the focal depth of the mask image, and the parallelism with the reference plane (in this case, the focal position of the mask image) is several μm. It is necessary to have an accuracy within As a method of obtaining a plane parallel to the reference plane, positioning is performed so that the plane connecting the tips of the three positioning pins 6 erected at three points on the plate 9 is parallel to the reference plane as shown in Fig. 2. Conventionally, a method has been used in which the pin 6 is adjusted and the tip of the pin is brought into contact with a plane of a target object, for example, the surface of the wafer 1, so that the plane is parallel to a reference plane.

A conventional plane positioning device using this method is
As shown in FIG. 3, screw holes e are provided in the plate 9 at approximately three equal points on one circumference, and positioning pins 6 are screwed into the screw holes e.
When the positioning pin 6 is turned, the tip of the pin 6 moves relative to the plate 9, so adjust the position of each pin 6 so that the plane connecting the tips of the three pins 6 is parallel to the reference plane Q. .

However, in this apparatus, since the wafer 1, which is the object, is pressed against a fixing pin provided on a reference surface, there is a drawback that if the pressing force is too strong, the wafer 1, which is the product, will be damaged. Furthermore, the leveling of the positioning pin 6 must be within several μm, and a mechanism for precisely adjusting the rotation angle of the positioning pin 6 is also required. Furthermore, if dust or a foreign object of 10 to 20 μm is present in the spherical seat 2a of the chuck 2, leakage will occur and the vacuum suction force will decrease.
There is also a possibility that the plane positioning accuracy will deteriorate when moving by h to .

Next, an example of a conventional plane positioning device used in a photomask visual inspection apparatus that requires highly accurate automatic focusing will be described with reference to FIG.

A photomask 11, which is an object to be inspected, is placed on a Z stage 13 provided on an XY stage 12. An objective lens 14 is arranged above the photomask 11. The pattern on the photomask 11 is enlarged and projected by the objective lens 14, and the pattern sensor 1 installed above the objective lens 14
5. On the other hand, the design data of the circuit pattern of the photomask 11 is stored as an image signal, and the signal output read from the magnetic tape 16 and the output signal of the pattern sensor 15 are transmitted to the defect determination section 17.
are electrically connected to be input to the The objective lens 14 for pattern detection has a nozzle formed at its tip, and is configured to be able to supply air. This air is supplied to the gap detection section 1.
8. Detection of these gaps uses the principle of an air micrometer, and the gap detection section 18 is configured to detect the gap g by detecting the difference between the back pressure of the nozzle of the objective lens 14 and the air supply pressure. ing. A controller 19 that drives the Z stage 13 is connected to a gap detection section 18.

To inspect the photomask 11, the entire surface of the photomask is scanned in the order of the arrows in FIG. While the XY stage 12 is being scanned for inspection, the output signals of the pattern sensor 15 and the magnetic tape 16 are compared in the defect determination section 17, and if a difference is detected, it is determined that it is an abnormal part. During this inspection, the photomask 11 has undulations and
Since an error due to the linear movement of the stage 12 is also added, it is necessary to always keep the objective lens in focus. Moreover, with the recent miniaturization of LSI patterns, 1μm
It has become necessary to detect the following defects as well. For this reason, it is necessary to use a lens with high resolving power, for example, a high resolution lens with an NA of 0.9 and a depth of focus of 0.2 μm, and a high focusing accuracy of 0.1 μm is required. High precision focusing accuracy of 0.1μ as above
In order to maintain m, there is a limit to the stroke of the Z stage, which can make minute changes, and is limited to 50 to 100 μ. However, with a stroke of this size, in addition to waviness on the surface of the photomask and errors in the linear movement of the Errors gradually occur from the inspection starting point during the inspection, and errors that are greater than the stroke of the Z stage may occur. Therefore, before automatic focusing, it is necessary to perform rough focusing, that is, to align the photomask with the focal plane of the objective lens within several μm. An example of a plane positioning device conventionally used in this device is shown in FIG. A in this figure is a plan view, and B in this figure is a sectional front view. The rotating base 21 is placed on a base 23 via a ball 22. A slope 21a is provided on the rotating base 21 to provide a wedge effect. A roller 25 provided on the upper plate 24 is in contact with the slope 21a, and a spring 26 is provided to provide this contact. The upper plate 24 is the lower base 2
3 is elastically supported by plate springs 27 and 28 via plates 29 and 30, and positioned and supported so as to be movable in the direction of arrow Y. One end of the rotating base 21 is
A spring 34 is provided for contacting and providing this contact with a nut 33 which is threaded into a screw 32 connected to the motor 31. The rotating base 21 is 3
It is rotatably supported by ball bearings 35.

In order to position the photomask 11 on the focus plane R of the objective lens 14 with the above configuration, the gap detection section 18 converts the back pressure at the gap g into a voltage.
The controller 19 drives the motor 31 based on the change in voltage of the gap detection section 18 until a certain voltage value is reached. When the motor 31 rotates,
The feed screw 32 rotates, and the feed screw 32 is attached to the nut 3.
3, the rotation base 21 in contact with the nut 33 rotates using the ball bearing 35 as a rotation guide. Along with this, the rotating base 21
Since the upper slope 21a also rotates, the slope 21a
The roller 25 in contact with moves vertically.
In this way, the photomask 11 is moved vertically, and the approximate focal range of the objective lens 14 is adjusted by several μm.
When the position is reached, the rotation of the motor 31 is stopped and positioning is completed. In this method, the second
Since it is non-contact compared to FIGS. 3 and 3, it has the advantage of not damaging the product and improving the positioning accuracy, but it has the following disadvantages.

To adjust the height of the entire photomask at once,
If the surface holding the photomask is tilted, if dust is trapped between the photomask and the holding table, or if the thickness of the photomask is uneven, plane positioning errors will occur. Furthermore, since detection is performed at one location, the detection section is aligned with the focal plane of the objective lens with high precision within a few μm, but the alignment accuracy of other sections deteriorates, and it is impossible to reduce this error. , is not possible with this device. Also, when positioning, several μ
Since it is necessary to maintain a plane of m, the resolution of the positioning device is required to be approximately 1 μm, and a complicated and highly accurate vertical mechanism as shown in FIG. 5 is also required.

As described above, in order to improve reliability and precision in semiconductor exposure equipment and mask inspection equipment, it is an important technical issue to align the surface of a target object with high precision in a non-contact manner.

[Purpose of the invention]

The purpose of the present invention is to solve the above technical problems,
To provide an automatic focusing method and a device thereof, capable of optically automatically focusing the surface of a target object with high precision and performing highly reliable and highly accurate appearance inspection, exposure, etc. It is in.

[Summary of the invention]

In order to achieve the above object, the present invention irradiates the surface of a target object with light, and directs the light reflected from the surface of the target object to be imaged by an imaging optical system before and after a conjugate focal plane. Two contrast signals are obtained by receiving the light with two arranged photoelectric conversion means, and focusing is performed by moving the target object in the optical axis direction of the imaging optical system so that the two contrast signals are equal. This is an automatic focusing method characterized by:
The present invention also provides a light irradiation means for irradiating light onto the surface of a target object, an imaging optical system for irradiating light with the light irradiation means and forming an image of reflected light from the surface of the target object;
two photoelectric conversion means arranged before and after a conjugate focal plane so as to receive the light imaged by the imaging optical system and obtain two contrast signals; and a target object control means for performing focusing by moving the target object in the optical axis direction of the imaging optical system so that the contrast signals obtained are equal.

[Embodiments of the invention]

Next, an automatic focusing device according to the present invention and a plane positioning device using the automatic focusing device will be explained with reference to FIGS. 7 to 10.

7 and 8 are diagrams showing a schematic configuration of a planar positioning device using an automatic focusing device according to the present invention, FIG. 7 is a planar layout diagram, and FIG. 8 is a schematic cross-sectional front view. It is.

Above the object to be positioned (the wafer 1 in this embodiment), position detectors 36a, 36 are placed at roughly three equal points on the circumference, which is slightly smaller than the plane of the object.
b, 36c are arranged. Furthermore, minute displacement mechanisms 37a, 37b, and 37c are provided below the chuck 2 that holds the wafer 1, which can drive the chuck 2 up and down.

The above minute displacement mechanisms 37a, 37b, 37c
are position detectors 36a, 36b, 36c, respectively.
Arranged at three equal points in the circumferential direction to correspond to
And, via the detection unit 41 and the controller 42,
They are configured to be controlled in accordance with detection signals from position detectors 36a, 36b, and 36c, respectively.

The chuck 2 that supports the wafer 1 is a plate 39.
and is elastically supported on the base 38 via a leaf spring 40.

In order to perform plane positioning of the wafer 1 with the above configuration, each detector 36 at the reference plane G must be
The outputs of a, 36b, and 36c are input to the detection section. The detection unit 41 instructs the control to raise the minute displacement mechanisms 37a, 37b, and 37c when the output is smaller than the reference output, and to lower them when the output becomes larger than the reference output.

When the surface of the wafer 1 is now on the plane H, the outputs of the detectors 36a, 36b, and 36c are zero. At this time, the detection unit 41 detects the detectors 36a, 36b, 3
The command continues to be issued to the controller 42 to raise the minute drive mechanisms 37a, 37b, and 37c until the above reference output is detected from 6c. When wafer 1 gradually rises to surface I, detector 3
The output starts to be output from 6a, and the detector 37b compares this output with the reference output, and the minute drive mechanisms 36a, 36b, and 36c are further raised until the output becomes equal to the reference output. The wafer 1 is raised and the detector 36
When a becomes the reference output, the minute drive mechanism 37a stops. However, since the outputs of the detectors 36b and 36c are lower than the reference output, the minute drive mechanisms 37b and 37c continue to rise. At this time, if the wafer 1 comes too close to the detector 36a and the output becomes larger than the reference output, the minute drive mechanism 37a is lowered to a position corresponding to the reference output. In this way, each detector 36
By raising or lowering the minute drive mechanisms 37a, 37b, 37c until the outputs of a, 36b, 36c match the reference output, a highly accurate reference plane, G plane, can be obtained in a short time. As an example, if electric micrometers are used for the detectors 36a, 36b, and 36c and a drive mechanism of about 1 μm is used as the micro drive mechanism, the detection accuracy can be measured with a resolution of 0.1 μm, and the micro stage that is driven up and down can also be used for strokes of 1 to 1 μm. At about 2 mm, the linearity of the drive displacement is high and a highly accurate flat surface can be maintained. And 1μm
The measuring pressure of an electric micrometer that measures the displacement of
Since the force is extremely small at 0.1N, there is no damage to the wafer during detection.

In the embodiments described above, a contact type position detector is used, but a non-contact type position detector can also be used.

When a non-contact type position detector is used, there is no risk of scratching or staining the object due to contact.

One of the non-contact type position detectors is an air micro type detector, but if this is used for a wafer or the like, there is a risk of stirring up dust and contaminating the target object. The use of an optical or electromagnetic detector is suitable for locating objects such as wafers that are extremely sensitive to dust. For this reason, the present invention was developed by paying particular attention to the fact that the surfaces of semiconductor products such as wafers and masks are nearly mirror-like.

An embodiment of an automatic focusing device using an optical detector according to the present invention will be described below with reference to FIGS. 9 and 10. FIG. 9 is a diagram showing a case where the automatic focusing device according to the present invention is applied to a photomask inspection device. An objective lens 14 is disposed above the photomask 11, and the pattern on the photomask 11 is enlarged and projected by this objective lens 14, and an image is formed on a pattern sensor 15 disposed above the objective lens 14. The formed image is inspected for defects using the same inspection method as described in the prior art. Above the photomask 11 is an objective lens 14.
In addition, three objective lenses 43a, 43 with low magnification
b and 43c are arranged inside the outer periphery of the photomask. Now, for ease of explanation, the objective lens 43
One axis using a will be explained using FIG. The controller 42 for the minute drive mechanism is the eighth controller.
Let's use the same one shown in the figure.

This embodiment uses illumination light to detect the position of the surface of the target object (the surface of the photomask 11), projects the image of this illumination light onto the observed object, and detects the intensity of the reflected light. A sensor is used to detect the intensity of the light intensity due to the movement of the object to be observed, and alignment is performed. Objective lens 43a
Illumination light is input from above, an image S of this illumination light is created on the optical path _L1 , and this image S is passed through the objective lens 43a.
An image is formed on the photomask 11 by this. This statue S ₁
, the optical path L ₂ branched by the half mirror 44a
Then, it is further projected onto the optical path _L3 branched by the mirror 45a. This projected image is referred to as _S2 , and light intensity sensors 46a and 47a are provided in front and behind _S2 .
When the enlarged projected image of the pattern of the photomask 11 comes into focus on the pattern sensor 15 of FIG. 9, the above-mentioned illumination image _S2 is focused on the O' plane. now photomask 1
A light intensity sensor 47a is disposed on the in-focus plane J' of the J plane, which is l above the O plane of 1, and a light intensity sensor 46a is disposed on the in-focus plane K' of the K plane, which is l below the O' plane. There is. With the above configuration, when the photomask 11 is located on the O plane, the light intensity sensors 47a and 46a are adjusted so that they show the same output, and the difference in output is set to zero.

On the other hand, a minute displacement mechanism 37a that can move up and down with an accuracy of 1 μm or less is provided below the photomask 11.

Next, the minute displacement mechanism 37a will be briefly explained.
When the motor 48 rotates, the feed screw 51 rotates via the gears 49 and 50. Since the feed screw 51 is screwed into a nut 53 fixed to the base 52, the feed screw 51 moves in the left-right direction in the figure, and the moving table 54 is accordingly moved in the same direction (direction A). . The roller 55 in contact with the slope 54a moves in the vertical direction, and at this time the lever 56 swings about the intersection of the plate springs 57 and 58 as a fulcrum.
The steel ball 59 attached to the lever 56 is connected to the roller 55
It moves in the vertical direction (direction B) by C/d, which is the vertical movement amount of . The rotation angle of the motor 48 is 15°, the gear ratio of gears 49 and 50 is 2.5:1, and the feed screw 51
Assuming that the pitch is 1 mm, the slope of the slope 54a of the movable table 54 is 1/8.33, and the ratio of the positional relationship between the roller 55 and the steel ball 59 is 4:1, the vertical motion resolution of the mask holder 60 is as shown in the following equation. be.

7.5°/360°×1/2.5×1000μm×1/8.33×1/2
.5=0.5μm For example, when the drive mechanism is raised, the photomask 11
As the position approaches the J plane from the O plane, the output of the light intensity sensor 47a increases and the output of the light intensity sensor 46a decreases. Light intensity sensors 47a, 46a at this time
By calculating the difference between the outputs, the direction in which the photomask 11 is moving can be determined. Based on this output, the motor 48 of the minute drive mechanism is rotated, and the photomask 11 is rotated.
Move down. In this way, the light intensity sensor 4
If the photomask 11 is moved until the output difference between 7a and 46a becomes 0, it will come to match the focal plane of the objective lens 14 in FIG. Photomask 1
If this is done simultaneously at three locations on the object lens 14, the focal plane of the objective lens 14 can be positioned in a short time.

According to the plane positioning device of this embodiment, there is no risk of damaging the wafer or mask, and moreover,
This method has the advantage that highly accurate planar positioning can be ensured in a short time because the two locations are detected simultaneously.
In addition, since the detection points are driven independently for positioning, even if the thickness of the product is uneven or the flatness of the holder holding the product is poor, the flatness relative to the reference plane can be maintained with high precision. There is also.

〔Effect of the invention〕

As described in detail above, according to the present invention, it is possible to use the inclined portion of the contrast signal, and to optically and highly accurately autofocus the surface of a target object such as a wafer or mask without contact. This has the effect of allowing highly reliable and highly accurate visual inspection, exposure, etc. to be performed.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram of a semiconductor exposure device, FIGS. 2 and 3 are explanatory diagrams of the same operation, FIG. 4 is an explanatory diagram of an LSI inspection device, and FIG. 5 is a plan view for explaining scanning of a photomask. FIG. 6 is an explanatory diagram of a conventional plane positioning device, FIGS. 7 and 8 are diagrams showing a schematic configuration of a plane positioning device using an automatic focusing device according to the present invention, and FIGS. 9 and 10. The figures show an embodiment of an automatic focusing device according to the present invention, with FIG. 9 being a schematic diagram for explaining the configuration, and FIG. 10 being a diagram for explaining the operation. DESCRIPTION OF SYMBOLS 1...Wafer, 2...Chick, 11...Photomask, 36...Detector, 37...Minute displacement mechanism, 41...Detection section, 42...Controller, 4
3...Objective lens, 46...Light intensity sensor.

Claims (1)

[Claims] 1. Light is irradiated onto the surface of a target object, and the light reflected from the surface of the target object is imaged by an imaging optical system using two photoelectrons arranged before and after a conjugate focal plane. The light is received by the conversion means to obtain two contrast signals, and the two
An automatic focusing method characterized in that focusing is performed by moving the target object in the optical axis direction of the imaging optical system so that the two contrast signals are equal. 2. A light irradiation means for irradiating light onto the surface of a target object;
an imaging optical system that irradiates light with the light irradiation means and forms an image of the light reflected from the surface of the target object; and an imaging optical system that receives the light that is imaged by the imaging optical system and obtains two contrast signals. moving the target object in the optical axis direction of the imaging optical system so that two photoelectric conversion means arranged before and after a conjugate focal plane and contrast signals obtained from each of the two photoelectric conversion means are equal; 1. An automatic focusing device comprising: a target object control means for performing focusing.