JPH0367103A - Device and method for detecting position - Google Patents

Abstract

PURPOSE:To allow position detection with high accuracy by detecting the relative positions of 1st and 2nd objects by the respective results of the detection at least at the two wavelengths by detecting means. CONSTITUTION:The alignment pattern on a mask 1 surface is constituted of a grating lens 3a having a prescribed focal length and the luminous flux for alignment made incident diagonally on the mask 1 surface from an alignment head 6 is deflected in the normal direction of the mask 1 surface and is con densed to a prescribed position. The alignment pattern on a wafer 2 is the grating lens 4a of the pattern asymmetrical with the Z-axis and guides the convergent light transmitted through and diffracted by the lens 3a toward the head 6. The alignment luminous flux 10a receives the lens effect of the lens 4a at this time and is made incident on the detector 8 in the head 6. A command signal is sent to a stage driver 101 in order to move an X-Y stage 100 so as to register the mask 1 and the wafer 2 in accordance with the output of the detector 8 in a CPU 102.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a position detection device and method, for example, a high-precision position detection device applicable to alignment between a mask or reticle (hereinafter collectively referred to as mask) of a semiconductor exposure device and a wafer. It is about detection.

"Conventional technology" In exposure equipment for semiconductor manufacturing, relative alignment of the mask and wafer has traditionally been an important element for improving performance.Especially in alignment in recent exposure equipment, In order to increase the degree of integration of semiconductor devices, there is a need for alignment accuracy of, for example, submicron or less.

In many alignment devices, a so-called alignment pattern for alignment is provided on the mask and wafer surface.
Alignment of both is performed using the position information obtained from them. As an alignment method at this time, for example, the amount of deviation between both alignment patterns may be detected by performing image processing, or U.S. Pat.
No. 037969, No. 4514858, JP-A-56-1
As proposed in Publication No. 57033, a zone plate is used as an alignment pattern, a light beam is irradiated onto the zone plate, and the position of a converging point on a predetermined surface of the light beam emitted from the zone plate is detected at this time. ing.

In general, alignment methods using zone plates are
Compared to a method using a simple alignment pattern, this method has the advantage of being able to perform alignment with relatively high precision without being affected by defects in the alignment pattern.

FIG. 9 is a schematic diagram of a conventional alignment device using zone plates.

In the same figure, a parallel light beam emitted from a light source 72 passes through a half mirror 74 and is condensed at a condensing point 78 by a condensing lens 76, after which it is placed on a mask alignment pattern 68a on the surface of a mask 68 and a support base 62. The wafer alignment pattern 60a on the surface of the wafer 60 is irradiated. These alignment patterns 68a and 60a are composed of reflective zone plates, and each form a focal point on a plane perpendicular to the optical axis including the focal point 78. At this time, the amount of deviation of the focal point position on the plane is detected by guiding the light onto the detection surface 82 using the condensing lens 76 and the lens 80.

Based on the output signal from the detector 82, the control circuit 8
4 drives the drive circuit 64 to position the mask 68 relative to the wafer 60.

FIG. 10 is an explanatory diagram showing the imaging relationship between the light beams from the mask alignment pattern 68a and the wafer alignment pattern 60a shown in FIG. 9.

In the figure, a part of the light beam diverging from the condensing point 78 is diffracted by the mask alignment pattern 68a,
A focal point 78a indicating the mask position is formed near the focal point 78. Further, the other part of the light beam passes through the mask 68 as zero-order transmitted light and enters the wafer alignment pattern 60a on the surface of the wafer 60 without changing its wavefront. At this time, the light beam is diffracted by the wafer alignment pattern 60a, and then passes through the mask 68 again as zero-order transmitted light, condensing near the converging point 78 and focusing on the wafer position.
form.

In the figure, when the light beam diffracted by the wafer 60 forms a condensing point, the mask 68 simply acts as a transparent state.

The position of the focal point 78b due to the wafer alignment pattern 60a formed in this way is determined according to the displacement amount Δσ of the wafer 60 with respect to the mask 68 along a plane perpendicular to the optical axis including the focal point 78. It is formed as a deviation amount Δσ' corresponding to Δσ.

[Problem to be solved by the invention] In these old notes, the wafer surface is tilted with respect to the mask surface, the reference plane such as the mask holder surface in the semiconductor exposure equipment, and the ground plane of the exposure equipment. If there is a sensor, the center of gravity of the light beam incident on the sensor changes, resulting in an alignment error.

$111J is the detection surface 1 when the wafer 60 is tilted by θ
It is assumed that the alignment light beam that has passed through the mask is now incident on the wafer 60 as shown in the figure.

At this time, if the surface of the wafer is tilted by an angle θ on average at the location where the alignment mark 60 is located, the center of gravity of the light amount on the detection surface 127 is Pθ, and the focal point P when there is no tilt. . Therefore, it has moved by Δδθ. This can be expressed as Δδθ=bw−tan2θ.

Now, if Δδθ ==18,7X10"X2XIO'=3.74
It becomes μm.

That is, the positional deviation error is 3.74 μm, and it is no longer possible to align the mask and wafer with the above accuracy.

SUMMARY OF THE INVENTION In view of the above-mentioned drawbacks of the prior art, it is an object of the present invention to provide a position detection device and method that enable accurate relative position detection that is not affected by wafer surface inclination.

[Means and effects for solving the problems] The present invention irradiates a first object with a plurality of light beams of wavelengths or multiple light beams with different wavelengths, and the light beams are converged or diverged by both the first object and the second object. The multi-wavelength light flux or the multiple light fluxes with different wavelengths are detected, and the relative positions of the first object and the second object are detected from the respective detection results at at least two wavelengths.

According to the embodiments of the present invention, which will be described later, not only is it not affected by the inclination of the sea urchin/S surface, but also it is not affected by positional fluctuations of the light beam irradiation means or the detector. Other effects will become clear in the description of the embodiments below.

〔Example〕

FIG. 1 is a schematic diagram of an exposure apparatus for manufacturing semiconductor devices to which a position detection apparatus according to a first embodiment of the present invention is applied. In the figure, 1 is a mask, 2 is a wafer, 3a and 4a are masks 1,
These are first and second physical optical elements consisting of Fresnel zone plates (grating lenses) formed on the wafer 2. 5 is, for example, a wafer chuck, and wafer 1
is adsorbed. 6 is an alignment head, which houses various elements for alignment. E is an exposure area for transferring the circuit pattern on the mask onto the wafer. There is an exposure light source (not shown) above the exposure area E in the drawing.

8 is a detector such as a CCD line sensor, 9 is a detection surface of the detector 8, IO is a light source whose oscillation wavelength range is variable, for example, a semiconductor laser, 11 is a light projecting lens system for a collimator, 12
is a half mirror.

Reference numeral 100 denotes an XY stage, which moves the wafer attracted to the wafer chuck 5 in the XY directions.

101 is a stage driver, 102 is a CPU, and based on the output of the detector 8, the mask 1 and Ueno\2
A command signal is sent to the stage driver 101 in order to move the xY stage 100 so as to align the two.

The XY stage 100 also holds the wafer 4 in a predetermined position in the Z direction.

The XY stage 100 includes a piezo-driven precision wafer stage and a stepping motor-driven rough wafer stage, the stage driver 101 includes this piezo and a stepping motor, and the CPU 102 uses the piezo to move the wafer minutely. When moving a target a large distance, a command signal is sent to the stepping motor.

In this embodiment, a light beam emitted from a light source 10 having a variable oscillation wavelength range, for example, a semiconductor laser, is turned into a substantially parallel beam by a projection lens system 11, and is passed through a half mirror 12 to a Fresnel zone on a mask 1, which is a first object. A first physical optical element 3a consisting of a grating lens, which is part of the plate, is irradiated from an oblique direction.

The first physical optical element 3a has a condensing or diverging effect, and emits the transmitted light in the normal direction (-2 direction) of the mask surface as the first object 1, so that it emits the transmitted light in the normal direction (-2 direction) from the first physical optical element 3a. The light beam is made incident on a second physical optical element 4a, which is a second object 2 and is formed by a grating lens provided on the surface of the wafer 2, at a distance of

The second physical optical element 4a has a condensing or diverging function, and emits a light beam in the direction of the alignment head 6, passes through the half mirror 12, and then condenses it on the detection surface 9 of the detector 8. The irradiation area on the mask is made larger than the size of the first physical optical element 3a, so that the state of the emitted light flux does not change even if the position of the first physical optical element 3a is slightly shifted due to mask installation error.

Hereinafter, for convenience, the i-th physical optical element 3a will be referred to as a mask grating lens, and the second physical optical element 4a will be referred to as a wafer grating lens.

In this embodiment, the alignment pattern on one surface of the mask is composed of a grating lens with a predetermined focal length, and the alignment light beam obliquely incident on one surface of the mask from the alignment head 6 is directed to the normal of the one surface of the mask. direction (-2 direction) and a predetermined position (for example, Z=+27
6.0 μm).

In this example, the angle α of oblique incidence on one surface of the mask is 10<a<80 (deg) degree is preferred.

Further, the alignment pattern 4a on the wafer 2 is a grating lens with an asymmetric pattern with respect to the Z axis, designed to have a focal length of 278.78 μm, for example, and converging (diverging) light transmitted through and diffracted through the grating lens on the mask 1 surface. Light is guided toward the alignment head.

At this time, the alignment light beam 10a receives the lens action of the grating lens and enters the light receiver 8 in the alignment head 6. In the first embodiment, alignment is performed in the longitudinal direction (y direction) of the scribe line where the pattern exists.

Here, the wafer 2 is
When the grating lenses 3a and 4a of the mask and wafer cause a positional change in the y direction, the grating lenses 3a and 4a of the mask and wafer will be in the same state as if the lenses were misaligned with each other in the lens optical system.
The output angle of the output light flux varies. Therefore, the light beam incident position on the light receiving surface 9 moves in the y direction on the light receiving surface 9 by an amount corresponding to the relative displacement amount in the y direction between the mask and the wafer. Here, the detector 8 is a CCD line sensor, and the element arrangement direction on the detection surface 9 coincides with the y direction. In a range where the relative displacement between the master and the wafer 2 is not so large, the amount of movement of the spot in the y direction is proportional to the relative displacement between the mask and the wafer in the y direction.

Now, the mask and the wafer are shifted by Δσ in the y direction, and the beam from the wafer 2 is focused (or diverged) by the grating lens 3a of the mask and enters the wafer. When the distance is a1 and the distance from the wafer 2 to the detection surface 9 is b, the amount of gravity shift Δδ of the light beam on the light receiving surface 9 is Δδ = ΔσX (-+1)
...(a). In other words, the center of gravity shift amount Δδ is (b/
a+1) times. A=(b/a+1) is the magnification of the deviation of the center of gravity of the light beam relative to the amount of positional deviation. However, the wavelength of the luminous flux at this time is assumed to be λ.

For example, if a==0.5 mm and b=50 mm, the center of gravity shift amount Δδ is expanded by a factor of 101 according to equation (a). Here, the center of gravity of the luminous flux is defined as the center of gravity of the luminous flux, within the luminous flux detection plane, when the product of the position vector from this point of each point in the detection plane multiplied by the light intensity of that point is integrated over the entire cross section, the integral value becomes 0 vector. It refers to a point.

Here, if the focal length of the grating lens 3a of the mask 1 at the wavelength λ is f1, and the distance between the wafers 2 and 2 is set as f1, then a = f + g, and therefore, the deviation of the center of gravity of the light beam at the wavelength λ is The magnification A for the amount of positional deviation is. Next, if the wavelength of the light source is λ and the radius of the annular zone of the grating lens is rm (m is the annular zone number), then the relationship with the focal length f is n7TT-f=mλ, and from this, If the wavelength of the alignment light beam changes by Δλ, then the focal length f changes by Δf, which is expressed by λ.

At this time, the magnification of the amount of deviation of the center of gravity of the light beam relative to the amount of positional deviation is: (f+g)2 λ r10g λ It changes by ΔA represented by , and becomes a magnification of A+ΔA.

For example, A = 1.01, focal length r- at wavelength λ
o.

If 187mm and g=0.03mm, then Δ　λ Further, if -=0 and 1, ΔA=8.6.

λ Therefore, mask 1 and sea urchin/%2 have a predetermined positional deviation amount Δσ
By modulating the wavelength of the light source by, for example, 10% or selectively souffling, the center of gravity of the light beam on the detection surface 9 is determined as follows: Δδ′=ΔA·Δσ (b)=8.
It moves by Δδ', expressed as 6 Δ σ .

Therefore, when the mask l and the wafer 2 are in a state where there is no misalignment in the y direction, that is, when Δσ = 0, even if the wavelength changes, the detection surface 9
If the grating lenses of the mask and wafer are designed so that the center of gravity of the light beam at the top does not change (that is, Δδ' - 〇), Δδ' will be proportional to the amount of misalignment between the mask and the wafer. . Therefore, by setting the values of λ and Δλ9g in advance, finding the values of f and A, and calculating the value of ΔA, the value of Δδ' can be calculated from the detection result of the detector 8 using equation (b). Easy mask and wafer deviation Δσ by substitution
can be detected. Also, at this time, when the wavelength is varied by Δλ, the moving direction of the center of gravity of the light beam corresponds to the direction of displacement of the mask and wafer in the y direction, that is, the positive and negative of Δδ′, and if this correspondence is determined in advance, It is also possible to detect the direction in which the center of gravity of the light beam deviates from the direction of movement when the wavelength changes.

Here, Δδ' when Δσ = 0 does not necessarily have to be O, but if the absolute value of Δδ' when Δσ = O and the direction of movement of the center of gravity of the light beam when the wavelength fluctuates are obtained in advance, position deviation can be detected. By substituting the difference between the value of Δδ' when Δσ=0 and the value of Δδ′ when Δσ=0 into equation (b), the amount of mask and wafer deviation can be determined. IfLri41>-rcJThe first method is to use the relational expression of the shift amount Δδ' of the center of gravity position of the light beam when changing the wavelength with respect to the amount of positional deviation Δσ between the mask and the wafer, that is, (b
) equation is determined in advance, and a predetermined 2
The light beams of different wavelengths are sequentially emitted, the center of gravity of each wavelength is detected by the detector 8, and the amount of movement Δ of the center of gravity of the light beam is detected.
δ′ is calculated, and from this value of Δδ′, the amount of positional deviation Δσ between both objects is calculated using equation (b), and the first object or the second object is moved by an amount corresponding to the amount of positional deviation Δσ at that time. move it.

The second method is to sequentially emit light beams of two predetermined wavelengths from a light source at the time of position detection, and detect a direction that cancels out Δδ′ and positional deviation amount Δσ from the center of gravity of the light beam obtained by the receiver 8. The first object or the second object is moved by a predetermined amount commensurate with Δδ', and at that point, the light beams of two predetermined wavelengths are sequentially emitted again until Δσ falls within the allowable range. Repeat the movement.

The following two CPU positioning procedures are shown in Figure 2 (
Shown in 1) and (2).

By doing as in the above embodiment, there is no need to determine the reference point for detecting the position of the center of gravity of the alignment light beam on the detection surface, that is, the position of the center of gravity of the light beam on the light receiving surface when Δσ = 0. By modulating or selectively shifting , the amount of relative positional deviation between the objects to be measured can be easily detected.

Further, when the wafer 2 is tilted, the amount of movement of the center of gravity of the light beam on the detection surface due to this tilt does not change even if the wavelength of the light source is changed, so Δδ' remains unchanged.

This is the same even if there is a change in the position of the light beam irradiation means or the detection means. Therefore, by detecting Δδ', detection can be performed without being affected by the tilt of the wafer or changes in the position of the light beam irradiation means or the detector.

In this example, a semiconductor laser was used as a light source, and the oscillation wavelength was modulated by controlling the injection current.

As a result, the absolute value of the relative positional shift can be detected simply by detecting the variation in the center of gravity of one light beam, and the relative position between two light beams when two light beams with different wavelengths are generated from different light sources. There is no need to worry about misalignment, etc.

By moving the wafer, which is the second object, based on the positional deviation amount Δσ determined in this way, the first object and the second object can be positioned with high precision.

As described above, the alignment light beam 10a is transmitted and diffracted by the grating lens 3a on the mask 1, and reflected and diffracted by the grating lens 4a on the wafer 2, thereby reducing the optical axis shift between the mask and the grating lens on the wafer. The light is magnified n times by a grating lens system at a predetermined magnification that depends on the wavelength, and then enters the light receiving surface 9 in the alignment head 6 . The light receiver 8 detects the center of gravity of the light beam.

Here, the focal length of the grating lens is the gap between the mask and wafer during exposure and the positional deviation Δ at a given wavelength.
It is set in consideration of the magnification of the beam gravity center position shift Δδ with respect to σ.

For example, consider a proximity exposure system in which the position of the center of gravity of the light beam is detected on the light-receiving surface 9 by magnifying the positional deviation between the mask and the wafer by 100 times at a predetermined wavelength with an exposure gap of 30 μm.

Now, assume that the wavelength of the alignment light beam from the semiconductor laser is 0.83 μm. At this time, the alignment light flux passes through the projection lens system 11 in the alignment head 6, becomes a parallel light flux, and passes through the grating lens system consisting of two grating lenses when passing sequentially through the Ueno X2 and then the mask 1. Schematic diagrams of the refractive power arrangement of the system at this time are shown in FIGS. 3 and 4. Note that this figure shows a system in which the grating lens 4a of the wafer is replaced with a transmission type grating lens that is equivalent to reflection.

3 shows a case where the grating lens 4a on the wafer 2 has a positive refractive power and a grating lens 3a on the mask l has a negative refractive power. FIG. 4 shows a case where the grating lens 4a on the wafer 2 has a negative refractive power and a mask This is a case where the grating lens 3a on the top has a positive refractive power.

Here, the negative refractive power and the positive refractive power are determined by whether minus order diffracted light or positive order diffracted light is used.

In the figure, for example, the aperture of the grating lens 3a on the mask 1 is 300 μm, and the aperture of the grating lens 4a on the wafer 2 is 280 μm, and the positional misalignment (axis misalignment) between the mask and the wafer is magnified 100 times, and The arrangement and the focal length of each element were determined so that the center of gravity of the light beam would shift and, as a result, the diameter of the light beam on the light-receiving surface 9 (Airy disk e-2 diameter) would be about 200 μm.

Next, the optical shapes of the mask grating lens 3a and the wafer grating lens 4a in this embodiment will be explained.

First, the mask grating lens 3a is set so that a parallel light beam having a predetermined beam diameter enters at a predetermined angle and is condensed at a predetermined position. Generally, the pattern of a grating lens is an interference fringe pattern on the lens surface when a coherent light source is placed at each of the light source (object point) and image point.

Here, the origin is located at the center of the scribe line width, with the X axis in the scribe line direction, the Y axis in the width direction, and the two axes in the normal direction of the mask surface. A parallel beam of light that is incident at an angle of α to the normal line of the mask surface and whose projected component is orthogonal to the scribe line direction passes through the grating lens 3a, is diffracted, and is then imaged at the focal point (Xl+ Y l+ z,). The equation of the curve group of the grating lens is as follows, where the contour position of the grating is represented by
2 (1) where λ is the center wavelength of the used wavelength range of the alignment beam, and m is an integer.

Assuming that the principal ray is incident at an angle α, passes through the origin on the mask surface, and reaches the focal point (x++ YI+ Z+), the right side of equation (1) is determined by the value of m, which is the wavelength m/ Indicates that the double optical path length is long (short), and the left side is the point (x + y, O) on the mask with respect to the optical path of the chief ray.
It represents the difference in the length of the optical path of a ray that passes through and reaches the point (X l+ V l+ Z'l).

On the other hand, the grating lens 4a on the wafer is set so as to condense a spherical wave emitted from a predetermined point light source onto a predetermined position (on the detection surface). The position of the point light source is determined by (x, , Y+, zl), where g is the gap between the mask and wafer during exposure.
g). The mask and wafer alignment is y
Assuming that the alignment is performed in the axial direction, and that the alignment beam is focused at the point (X2+72+z2) on the detection surface when the alignment is completed, the equation of the curve group of the grating lens on the wafer is expressed in the coordinate system defined earlier. 11X2+Y2+22 r70mλ/2...
(2) It is expressed as

Equation (2) shows that the wafer surface is at z=-g, the chief ray is at the origin on the mask surface, the point on the wafer surface (0, 0, -g), and the second point on the detection surface (X21 y21 22) Assuming that the ray passes through the grating (X+Y+3
) is an equation that satisfies the condition that the difference in optical path length between the ray passing through and the principal ray is an integral multiple of a half wavelength.

Generally, a zone plate (grating lens) for a mask is a 0, 1 amplitude type grating in which two areas are alternately formed: a region through which light rays pass (transparent part) and a region through which light rays do not pass (shade part). Created as an element. Further, a zone plate for a wafer is made, for example, as a phase grating pattern with a rectangular cross section. (1), (2)
In the formula, the outline of the grating is defined at a position that is an integer multiple of a half wavelength with respect to the principal ray, which means that in the grating lens 3a on the mask, the line width ratio of the transparent part and the light shielding part is 1:1. In the grating lens 4a on the wafer, the ratio of the lines and spaces of the rectangular grating is 1:l.
It means that.

The grating lens 3a on the mask is formed, for example, by transferring a grating lens pattern of a reticle previously formed by EB exposure onto an organic thin film made of polyimide, or the grating lens on a wafer is formed after forming the exposure pattern of the wafer on the mask. It is formed by exposure transfer.

FIG. 5(A) shows a grating lens 4a on the wafer surface.
, the grating lens 3a on the mask surface is shown in FIG.
・A pattern of one example is shown.

Next, a case will be specifically described in which a predetermined positional shift amount is given between the mask and the wafer in the embodiment shown in FIG. 1(A).

First, a semiconductor laser (wavelength 8
:30 nm) passes through the projection lens system 11 and becomes a parallel light beam with a half-width of 600 μm.
incident at degrees.

The scribe line on the mask surface has a width of 60 μm and a length of 2.
A grating lens 3a of 80 μm is set, and a grating lens 4a of the same size is set on the scribe line on the wafer surface. The relative positional deviation between the mask and the wafer is controlled by a precision wafer stage driven by a piezo to provide a small amount of displacement, and a coarse stage for the wafer driven by a stepping motor to provide a relatively large amount of displacement. or,
The amount of displacement was measured using a sub-length machine (resolution: 0.001 μm) in a constant temperature chamber at a controlled temperature of 23° C.±0.5° C.

Further, a one-dimensional CCD line sensor was used as a light receiver for detecting the center of gravity position of the light beam within the alignment head 6.

The element arrangement direction of the line sensor matches the positional deviation detection direction (alignment direction). The output of the line sensor is subjected to signal processing so as to be normalized by the total light intensity of the light receiving area. As a result, even if the output of the alignment light source varies somewhat, the measured value output from the line sensor system accurately indicates the position of the center of gravity.

The resolution of the center of gravity position of the line sensor was 0.2 μm when measured with a 50 mW semiconductor laser, although it depends on the power of the alignment light beam.

Grating lens 3 for a mask according to the first embodiment
In the design example of grating lens 4a for grating lens 4a and wafer, the positional deviation between the mask and wafer is magnified 100 times, and the center of gravity of the signal light flux is set to move on the sensor surface.

Here, assuming that the amount of misalignment between the mask and the wafer is 3.0 μm, approximately 30011. The center of gravity of the light quantity will be at the position m.

Since the position of the center of gravity of the light beam on the sensor when the positional deviation amount is 0.0 μm is unknown, by changing the semiconductor laser injection current in the alignment head by 50 mA, the lateral deviation detection magnification variation width ΔA is changed from 830 nm to 838 nm by changing the oscillation center wavelength from 830 nm to 838 nm. 0.82, and the shift amount of the center of gravity of the light amount on the line sensor is 2.48 μm.

Furthermore, when the amount of misalignment between the mask and the wafer is 7.0 μm, the amount of light intensity shift on the sensor at the same wavelength modulation rate is 5.74.
It becomes μm.

In this way, assuming that the wavelength modulation rate of the light source is constant, the amount of movement Δδ of the center of gravity of the light amount on the sensor before and after the modulation of the oscillation wavelength is
' to 1. Relative positional deviation amount Δσ between mask and ueno
is calculated as shown below.

FIG. 6 shows the change in the center of gravity position detected by the alignment line sensor when a predetermined amount of positional deviation is actually applied between the mask and the wafer. As is clear from FIG. 6, the detected gravity center position has a linear relationship with the amount of positional deviation between the mask and the wafer, with the magnification of the grating lens system being a proportionality constant. However, linearity is determined when the amount of positional deviation is a constant value (20
μm), this no longer holds true and nonlinearity appears.

This is because as the amount of axial misalignment between the mask and the grating lens on the wafer increases, the wavefront aberration of the light beam becomes significant, and asymmetry appears in the spot shape on the sensor.

This wavefront aberration becomes more obvious as the NA of the grating lens increases. Therefore, when setting a grating lens in a fixed area, it is desirable to make the NA as small as possible.

In the positioning device in this embodiment, the resolution of positional deviation is 0.002 μm 1 20mm above the positional deviation measurement range.
μm (linear region) is obtained.

In this embodiment, since the light beam is obliquely incident on the mask surface and the oblique receiving optical path is set, it is possible to measure and control the amount of positional deviation between the mask and the wafer without the alignment head 6 entering the exposure area E. can.

FIG. 7 shows a perspective view of a main part of a semiconductor exposure apparatus to which a second embodiment of the invention is applied. Components that are the same as those in FIG. 1 are designated by the same reference numerals. The main components are the same as in the first embodiment, but in this embodiment, two semiconductor lasers IQ-1, 1 with different oscillation center wavelengths are installed as light sources in the alignment head.
0-2 was set. The optical paths of the light beams from the semiconductor lasers 10-1 and 10-2 are adjusted so that their principal rays overlap when emitted by the half mirror 12a.

The CPU 102 executes the wavelength change described in the above embodiment. This may be achieved by providing a shutter in front of each semiconductor laser and controlling the shutters to open and close alternately.

Assuming that the center wavelength of the semiconductor laser 10-1 is 830 nm and the center wavelength of the semiconductor laser 0-2 is 780 nm, and that the grating lens design equation is applied to alignment light with a wavelength of 805 nm, the Balano 6.2X10-
2 Focal length f1 of grating lens 3a of mask l Mask, wafer spacing g are the same as in the first embodiment, wavelength 80
When the positional deviation detection magnification A of the grating lens system approaches lOO for a 5 nm light beam, the magnification variation width Δ
A becomes 5.33.

FIG. 8 shows a perspective view of essential parts of a semiconductor exposure apparatus to which a third embodiment of the present invention is applied. In this embodiment, as a light source,
Using a white light source 10', a diffraction grating 1 is used as a wavelength selection means.
3 and a slit plate I4 are set within the alignment head 6. The diffraction grating 13 changes the angle of incidence of the diffracted light for each wavelength by changing the angle of incidence of the incident light beam. A slit plate I4 with slits, pinholes, etc. is placed at a predetermined position where it can receive diffracted light, and the diffraction grating 1
By rotating the light beam 3 so that the angle of incidence changes, the wavelength of the light beam passing through the slit plate 14 can be changed. Therefore, when changing the wavelength as described in the above embodiment, the CPU 102 changes the wavelength of the irradiation light beam by rotating the diffraction grating 13 by a predetermined angle.

As in this example, by using a light source with low coherency, the resist surface roughness on the two sides of the wafer can be masked and
Unnecessary light such as speckles on the light receiving surface 9 caused by factors such as scattered light from the edges of alignment marks on the wafer can be suppressed.

Note that the wavelength selection means is not limited to using a diffraction grating, and for example, a prism or the like may be used. Further, the wavelength selection means is arranged not immediately after the light source 10' but just before the detection surface 9 of the detector 8, and the grating lenses 3a and 4a are irradiated with white light, and the diffracted light beam is detected by the wavelength selection means. It may be controlled so that only the desired wavelength is incident on the detection surface 9. In this case, the slit plate 14
The size of the slit is set so that even if the emission angle of the light beam changes due to either the mask or the wafer, as long as the mask and wafer are within a predetermined misalignment range, the light of the detection wavelength will not be blocked. put.

A fourth embodiment according to the present invention is shown in FIG.

Like the previous embodiment, this embodiment is an alignment device between a mask (reticle) and a wafer of a semiconductor exposure apparatus, and a resist having a predetermined thickness is spin-coded on the wafer.

In this embodiment, the wavelength of the alignment light beam is selected based on the film thickness of the resist and the spectral reflectance, so that the intensity level of the light beam focused on the sensor surface is always maintained above a certain level.

In this embodiment, the spectral reflectance determined by the resist film thickness is measured by optical means prior to positional deviation amount measurement control.

The optical means for measuring spectral reflectance includes an optical head 6 for measuring positional deviation, a sensor 13 attached to the main body of the exposure apparatus,
Using the mirror 14, the mask 1 is placed in the exposure area E.
Before mounting on the wafer (or after mounting), a light beam 10a is projected onto the wafer 2 surface from the light source 10 in the head 6, and the intensity of the reflected light Job from the wafer surface on the sensor 13 is measured.

The light source 10 includes a plurality of lasers with different oscillation wavelengths, a tunable laser with a controllable oscillation wavelength (semiconductor laser, dye laser, etc.), a quasi-monochromatic superluminescent diode (SLD), or a white light source and wavelength selection means ( It consists of an optical means consisting of a prism or a diffraction grating, which scans wavelengths in a selectable wavelength range to measure the spectral reflectance from the wafer surface coated with resist, and then measures positional deviation based on the measured data. At least 2 to choose from
Determine the two wavelengths.

In the system shown in FIG. 12, the wavelength selection method is as follows based on the beam projection angle α and the incident angle α' of the signal beam 10a diffracted from the mark 3a on the mask surface and reaching the wafer 2 surface. Decide as follows.

In general, film thickness l formed on a substrate with refractive index no, refractive index n
The intensity ■ of a light beam incident on the film thickness at an angle θ and the reflection intensity In of a light beam of wavelength λ are given by the following equation, taking into account multiple reflections.

λ22, r represents the amplitude reflectance at the interface between the thin film and the substrate, and r' represents the amplitude reflectance at the second interface between the thin film and the atmosphere, etc.

Since 1r becomes maximum when δ becomes δ=(2m-1) yr (m: integer), when the following is satisfied, the reflectance considering multiple reflections becomes maximum.

The wavelength that gives the maximum reflectance corresponding to the incident angle θ is now λ
(θ). Therefore, the wavelength λ(θ') that provides the maximum reflectance corresponding to other incident angles θ' is given by using λ(θ). In general, the dependence of the wavelength on the incident angle θ such that the reflectance is at a predetermined ratio with respect to the maximum value is also given by equation (3).

Therefore, by measuring the spectral reflection characteristics at the incident angle α 2 in advance, it is possible to determine the wavelength at which the reflectance is at a predetermined ratio with respect to the maximum value based on equation (3).

Further, the wavelength at which the reflectance is selected as a percentage of the maximum can be set arbitrarily, and may be determined by taking into consideration the selectable wavelength range of the light source.

Incidentally, the spectral intensity characteristics of the signal light flux on the sensor according to the resist film thickness may be measured using a positional deviation measuring optical system that includes the optical head 6, the mask 1, and the wafer 2.

Note that, as in the previous embodiment, in this embodiment, light beams of two different wavelengths are projected onto a grating lens formed on a mask and a wafer, and the amount of positional deviation is measured by utilizing the difference in positional deviation detection sensitivity depending on the wavelength. The principle, signal processing, etc. are as described above.

〔Effect of the invention〕

As explained above, according to the present invention, even if a second object such as a wafer is tilted or a positional change occurs in the light beam irradiation means or the detector, highly accurate position detection without causing errors is possible. .

[Brief explanation of drawings]

FIG. 1 is a schematic diagram of a first embodiment according to the present invention, FIG. 2 is a flowchart showing an embodiment of the alignment control procedure of the present invention, FIGS. 3 and 4 are explanatory diagrams of grating lens refractive power arrangement, Figure 5 is an example of an alignment mark pattern, Figure 6 is an example of an alignment mark pattern.
The figures are graphs showing the positional deviation detection characteristics of the present invention, Figure 7 is a schematic diagram of the second embodiment of the present invention, Figure 8 is a schematic diagram of the third embodiment of the present invention, Figures 9 and 7 are graphs showing the positional deviation detection characteristics of the present invention. Figures 10 and 11
The figure is an explanatory diagram of a conventional example, and FIG. 12 is a schematic diagram of a fourth embodiment according to the present invention. In the figure, 1: mask, 2: wafer, 3a, 4a: grating lens, 8: detector, 9: detection surface, 10: light source 10
2: CPU. 5th □□□ (A) (Usuun

Claims (2)

[Claims] (1) A device for detecting the relative position of a first object provided with a first physical optical element having a lens function and a second object provided with a second physical optical element having a lens function, a light source means for emitting light of a plurality of wavelengths simultaneously or sequentially; a detection means for detecting light of a plurality of wavelengths, and the relative position of the first object and the second object is detected based on the detection results of each of the at least two wavelengths by the detection means. Device. (2) A method of detecting the relative position of a first object and a second object, in which a first physical optical element having a lens effect is attached to the first object, and a second physical optical element having a lens effect is attached to the second object. and irradiate the first physical optical element with light of a plurality of wavelengths simultaneously or sequentially, and the light is focused or diverged by the first physical optical element and the light is focused or diverged by the second physical optical element. A position detection method, comprising: detecting the plurality of wavelengths of light; and detecting the relative positions of the first and second objects based on the detection results of at least two wavelengths.