JPH0371002A - Positioning device - Google Patents

Abstract

PURPOSE:To position two objects with high accuracy by finding the quantities of relative displacement of both the objects to be positioned from the phase difference between the beat signals generated with reflected light beams detected by 1st and 2nd photodetectors. CONSTITUTION:Monochromatic light from a light source is multiplexed by a flange light generating means while having a specific optical path difference and a specific frequency difference to generate fringe light where fringes having specific intervals run at a beat frequency. This fringe light irradiates positioning marks formed on the surfaces of two objects to be positioned by mirrors 36 and 42. Then light such as scattered light and diffracted light which contains position information is reflected by each positioning mark and reflected light beams are converted photoelectrically by the photodetectors 50 and 51 and outputted as beat signals. Then a signal processing circuit 52 finds the phase difference between the two beat signals to find the quantities of relative displacement of the two objects, which are positioned according to the quantities.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention provides a suitable position for aligning two alignment objects, such as a wafer and a reticle, in an exposure apparatus or the like for manufacturing semiconductor ICs or LSIs. Regarding a matching device.

[Conventional technology]

With the miniaturization of semiconductor ICs and LSIs, it is necessary to align the relative positions of the mask and wafer with high precision in equipment that exposes and transfers mask patterns to wafers in a batch manner or in a step-and-repeat manner. be.

As a method for performing precise alignment using diffracted light, there is a method using diffracted light from a large number of gratings, as disclosed in Japanese Patent Application Laid-Open No. 60-256002.

FIG. 7 is a diagram showing an alignment device to which the above alignment method is applied. This device expands a laser beam output from a laser light source 1 to a predetermined beam diameter with a beam expander 2,
After shaping the beam into an elliptical beam with an elongated cross-sectional shape using a cylindrical lens 3, the beam is reflected by a mirror 4 and is reflected by a lens 5. The beam is guided to a mirror 8 via a beam splitter 6 and a lens 7.

Further, the laser beam guided by the mirror 8 is once focused into a slit shape and then reflected by a mirror 10 which is placed below the reticle 9 and has a reflective surface parallel to the reticle 9. The laser beam reflected by the mirror 10 is further reflected and guided to the entrance pupil 12 of the reduction projection lens 11, and the wafer stage 1
A long and thin strip-shaped spot light LA adjusted by the cylindrical lens 3 is projected onto the wafer 14 held on the wafer 3.
is imaged.

Here, as shown in FIG. 8, on the wafer 14, mark rows M1 to M3, which are formed by arranging a plurality of single marks at predetermined intervals on the same straight line, are arranged perpendicularly to the arrangement direction of the single marks. A plurality of alignment marks arranged in a plurality of rows at equal intervals in the direction are formed in the shape of a diffraction grating.

In this way, since the alignment mark has a diffraction grating shape, when the spot light LA is irradiated onto the alignment mark, scattered light is emitted from the top surface of the wafer.

Reflected light such as diffracted light is generated. The reflected light containing the information of this alignment mark (position information of the wafer 14) enters the reduction projection lens 11 in the opposite direction, passes through the entrance pupil 12, and then passes through the mirror 10.
．． It is reflected by the mirror 8, reaches the beam splitter 6 via the lens 7, is further reflected by the beam splitter 6, and is guided to the spatial filter 15. Of the reflected light from the surface of the wafer 14, the scattered light (first-order or higher-order diffracted light) is displaced with respect to the optical path of the specularly reflected light by an amount higher in spatial frequency than the specularly reflected light (0th-order diffracted light). Therefore, entrance pupil 1
Of the reflected light that has reached the spatial filter 15, which is conjugate with 2, the specularly reflected light is blocked, and only the scattered light (first-order or higher diffraction light) passes through. Scattered light 9 that passed through the spatial filter 15
The diffracted light is focused onto a photodetector 17 by a focusing lens 16 . The twice-diffracted scattered light focused on the photodetector 17 is photoelectrically converted and output as a photoelectric signal SA corresponding to its intensity. The photoelectric signal SA output from the photodetector 17 is input to the alignment processing circuit 18. Also, at this time, the positional information of the alignment mark on the wafer 14 corresponding to the photoelectric signal SA is measured by one length measuring device, and is sent to the alignment processing circuit 18 as a time-series up-down pulse, for example.
is man-powered.

In such an apparatus, when the wafer stage 13 is moved by a drive system (not shown) and the alignment mark is scanned in the direction of the arrow with the spot light LA shown in FIG.
A photoelectric signal SA corresponding to the intensity of diffracted light at each position of the alignment mark and position information PD of the wafer 10 are input to the alignment processing circuit 18 .

FIG. 9 is a diagram showing a graph created from the photoelectric signal SA obtained by scanning the alignment mark and the position information PD. If alignment marks on the wafer 14 are provided for each chip on the wafer 14, the peal positions PL, P2 . By detecting the average value of P3, the center of the chip and the spot light L
The center of the optical axis of A can be aligned.

[Problem to be solved by the invention]

The conventional alignment apparatus described above is suitable for aligning the center of the laser beam and the wafer 14, but cannot directly align the wafer 14 and the reticle 9. Therefore, it is necessary to align the reticle 9 with the center of the laser beam in advance, and a separate optical system is required for this purpose, making the device larger and more complex, and the accuracy of the separately sold optical system directly affects the alignment accuracy. There were problems with the impact.

Furthermore, since the positional relationship between the wafer 14 and the reticle 9 is not constantly measured, even if the position of the reticle 9 deviates from the initially set position due to unforeseen circumstances, the operator may not be able to notice the positioning. There is a possibility that it will continue.

Naturally, in such a case, there is a problem that accurate positioning cannot be performed.

Furthermore, since alignment is performed using the peak value of the intensity of the diffracted light, there is a problem that it is susceptible to changes in the intensity of the diffracted light and noise.

Furthermore, the wafer stage 13 is moved and the wafer 1
Since the positional information of 4 is obtained, time unrelated to direct alignment is consumed, resulting in poor work efficiency, and moreover, alignment cannot be performed with accuracy higher than the length measurement resolution of the wafer stage 13. There is a problem.

Therefore, the first object of the present invention is to provide an alignment device that can align two alignment objects with high precision, speed up the alignment time, and simplify the alignment optical system. There is a particular thing.

[Means to solve the problem]

In order to solve the above problems, the present invention combines a light source that generates monochromatic light and i41 color light outputted from this light source with a predetermined optical path difference and frequency difference to generate fringe light flowing at a heat frequency. a fringe light generation means; a fringe light irradiation means for irradiating the fringe light generated by the fringe light generation means onto alignment marks formed on two alignment objects; first and second light detection means each detecting reflected light including a positional information component such as scattered light and double-diffraction light reflected from each alignment mark by irradiating the alignment mark with the fringe light; Based on the phase difference between the heat signal generated from the reflected light detected by the first light detection means and the heat signal generated from the opposite light detected by the second light detection means, the two alignment targets are determined. The structure includes a signal processing means for determining the relative displacement of an object.

[Effect]

By taking the above measures, the present invention combines monochromatic light output from a light source with a predetermined optical path difference and frequency difference in the fringe generation means, and generates fringe light in which fringes at predetermined intervals flow at a heat frequency. be done. This fringe light is irradiated by the fringe light irradiation means onto the alignment marks formed on the surfaces of the two alignment objects, respectively. Then, light such as scattered light and diffracted light containing positional information from each alignment mark is reflected, and the reflected light is photoelectrically converted by the first and second photodetecting means and output as a heat signal. . Then, the signal processing means determines the phase difference between the two heat signals, the relative displacement of the two alignment objects is determined from this phase difference, and the alignment of both alignment objects is determined based on this displacement amount. will be carried out.

〔Example〕 Hereinafter, one embodiment of the present invention will be described with reference to the drawings.

The alignment apparatus according to this embodiment includes a beam irradiation system as a fringe light irradiation means, a signal detection system, and a fringe beam generation system, and is used to align a wafer and a reticle.

FIG. 1 is a diagram showing a beam irradiation system and a signal detection system.

The beam irradiation system shown in the figure includes a wafer beam irradiation system that irradiates the wafer 20 with a fringe beam and reads its position information, and a reticle beam irradiation system that irradiates the reticle 21 with a fringe beam and reads its position information. Consisting of The beam irradiation system for the wafer includes a reduction projection lens 3 which is disposed between the wafer 20 and the reticle 21 which are arranged facing each other and has an incidence @30, and a polarization plane of a fringe beam sent from a fringe beam generation system which will be described later. a 45-degree rotator 32 that rotates the fringe beam by 45 degrees, a polarizing beam splitter (hereinafter referred to as PBS) 33 that reflects the S-polarized component of the fringe beam and transmits the P-polarized component, and a λ/ placed on the same optical axis
4 boards 34. A lens 35, a mirror 36, and a lens 37 arranged on the same optical axis on the reflection side of the reflected light incident on the beam splitter 33 from the wafer 20 side. Spatial filter 38
, and a lens 39. Furthermore, the reticle beam irradiation system uses a 45 degree rotator 32. which is shared with the above beam irradiation system. PBS33 and PB84 that reflects the S polarized light component divided by this PB833 to the reticle 21 side.
0, and a λ/4 plate 41.0 disposed on the reflection side of the S-polarized light component of this PBS0 40.
A mirror 42 and a lens 43 disposed on the same optical axis on the side through which the reflected light from the reticle 21 of the PBS 40 is transmitted. Spatial filter 44. It is composed of a lens 45.

The signal detection system includes a first photodetector 50 that photoelectrically converts reflected light collected by a lens 39 and outputs a heat signal, and a first photodetector 50 that photoelectrically converts reflected light collected by a lens 45 and outputs a heat signal. A signal processing circuit 52 takes in the heat signals output from the photodetectors 50 and 51 and calculates a phase difference.

The beam irradiation system and the signal detection system constitute a set of alignment optical systems. In reality, alignment is performed by arranging a plurality of sets of such alignment optical systems.

Here, the reduction projection lens 31 exposes an image of a circuit pattern or the like formed on the reticle 21 onto the wafer 20 after reducing the image to 115 or 1/10.

Note that a spectroscopic means such as a beam splitter can be used in place of the PB833 and the 45-degree rotator 32. Also, PBS40. The λ/4 plate 41 may also be a simple spectroscopic means, and in this case, the λ/4 plate 41 is not necessary. The reticle 21 is mounted on a reticle stage (not shown), and this reticle stage is finely driven in the X, Y directions, and the θ (rotation) direction by a driving means (not shown). Then, the reticle 21 is positioned at a predetermined position, for example, with respect to the optical axis of the reduction projection lens 31 using a positioning mechanism (not shown). This alignment mechanism is capable of fine movement of, for example, about 2 nm. Further, the reticle 21 is attached to the wafer 20.
It is illuminated with light of a wavelength effective for exposing the resist coated thereon (for example, exposure light including g-line, i-line, etc.). That is, the light beam forming the pattern image of the reticle 21 is imaged on the surface of the wafer 20 by exposure illumination. On the other hand, the wafer 20 is
It is placed on a Y stage and is movable in two dimensions, the X direction and the Y direction. These two XY stages operate in the X direction, Y direction, and Z direction (optical axis direction).
and a fine drive mechanism (motor, etc.) in the θ direction (rotation direction). The position coordinates on the XY stage are constantly detected with an accuracy of, for example, about 0.1 μm by a length measuring means such as a laser interferometer. Incidentally, the wafer 20 and the reticle 21 are formed with alignment marks in which single marks are arranged in a matrix as shown in FIG.

This alignment mark is a diffraction grating-like mark row AMI
~AM5 are arranged at constant intervals LO.

FIG. 3 is a diagram showing a fringe beam generation system.

This fringe beam generation system includes a laser light source 61 (hereinafter referred to as a "light source") that outputs linearly polarized monochromatic light;
An acousto-optic element 62 (hereinafter referred to as an "AO element") that diffracts a part of the laser light output from the light source 61 at a predetermined angle θ and provides a frequency difference f with respect to the non-diffracted light.
) and an afocal lens 6 that transmits the light source wavefront a of this beam without changing it to a different wavefront.
3 and a 45 degree opening that rotates the polarization plane of this beam by 45 degrees.
A PBS 65 that splits the light into a polarized light component and an S polarized light component, and a beam splitter 66 arranged on the side that reflects the S polarized light component.
An afocal magnifying lens 67 images the light reflected by the beam splitter 66 on the photoelectric detector 68 and 69, and the laser beam magnified by the afocal magnifying lens 67 forms an image. A photoelectric detector 68,69 that photoelectrically converts the image, a signal processing circuit 70 that calculates the phase difference between the detection signals output from the photoelectric detector 6869, and a signal processing circuit 70 that calculates the phase difference between the detection signals output from the photoelectric detector 6869. and a high frequency power source 71 that changes the frequency applied to the AO element 62 based on the frequency applied to the AO element 62.

In the fringe beam generation system configured in this way, a linearly polarized laser beam emitted from the light source 61 is transmitted to the AO element 62.
A part of the incident beam undergoes a Doppler shift of frequency fs and is diffracted at an angle θ. here,
When a plane wave beam parallel to the optical axis and a plane wave beam at an angle of 4 to the optical axis are emitted from the light source wavefront a, there is a frequency difference fs between the two beams, so in an afocal system, the conjugate of the virtual light source wavefront If a plane is taken, interference light flowing at a frequency fs can be generated. Therefore, the beam emitted from the AO element 62 passes through an afocal lens 63 that faces plane a and whose conjugate plane is c, and its polarization plane is rotated by 45 degrees by the 45-degree low chitaro 4, resulting in P B S 65 i,1 As a general rule, the P-polarized reflected light component from the PBs 65 becomes a fringe beam once.

This fringe beam is guided by an afocal system (not shown) to a beam irradiation system Lx for positioning in the X direction. On the other hand, the S-polarized reflected light component from the PBS 65 once becomes a fringe beam at the surface C, and is guided to a beam irradiation system Ly for positioning in the Y direction by an afocal system (not shown). The beam reflected by the beam splitter 66 is imaged on a surface d by an afocal magnifying lens 67. Photoelectric detectors 68 and 69 are arranged at a predetermined distance from each other on this imaging plane, and the beams incident on the detection plane of each photoelectric detector 68 and 69 are photoelectrically converted into electric signals Φα and Φβ, respectively. Ru. These detection signals Φα and Φβ are manually input to the signal processing circuit 70. In the signal processing circuit 70, the detection signal Φ
The phase difference between α and Φ3 is detected, and based on the detection result, a control signal set so that the Tachikawa difference becomes a predetermined phase difference is output to the high frequency power supply 7. High frequency The power supply 71 controls the frequency applied to the AO element 62 based on the input control signal.The AO element 62 changes the diffraction angle θ of the incident laser beam depending on the applied frequency.

Then, the fringe beam sent from the fringe beam generating yarn is rotated by a 45 degree rotator 32 to change the polarization plane to 45 degrees.
After being turned around, it enters PB833. PBS
The fringe beam incident on 33 is divided into a P polarization component and an S polarization component.

Among these divided polarized light components, the P polarized light component fringe beam passes through the PBS 33 and passes through the λ/4 plate 34 into a circle < Gj
The light becomes light, passes through the lens 35, and reaches the mirror 36. The fringe beam reflected by the mirror 36 is focused by -/ffi, then reflected by the lower surface of the reticle 2 and guided to the entrance pupil 30 of the reduction projection lens 31. The fringe beam that enters the entrance pupil 30 and passes through the reduction projection lens 31 passes through the lens 35 and the reduction projection lens 31.
As a result, an image is formed on the wafer 20 with a beam profile having fringes with a spacing PO as shown in FIG. Note that the beam profile imaged on the wafer 20 is an instantaneous image flowing at the bit frequency, and if the bit frequency is several Mllz, for example, it cannot be confirmed visually. Since the fringe imaged on the wafer 20 flows at the heat frequency, each mark row AMI to AM5 of the alignment marks formed on the wafer 2 is scanned. The alignment marks scanned by the fringe beam have a diffraction grating pattern as shown in Figure 5, so each mark row emits specularly reflected light (0th order diffracted light) and diffracted light of the fringe beam. (diffraction light of first order or higher order) is reflected as reflected light. The reflected light containing the positional information of the wafer 20 from these alignment marks is transmitted to the reduction projection lens 3
It is inversely incident on 1. The reflected light that entered the reduction projection lens 31 in the opposite direction passes through the entrance pupil 3o, is reflected by the lower surface of the reticle 21 and the mirror 36, and is further reflected by the lens 35. The light passes through the λ/4 plate 34 and enters the PB 533. The reflected light from the wafer 20 that is reversely incident on the PB 333 is reflected by the PB 833 and reaches the spatial filter 38 via the lens 37 . The spatial filter 38 is a reduction projection lens 31. The lens 35 and the lens 37 form a Fourier plane with respect to the surface of the wafer 20, and only the specularly reflected light from the surface of the wafer 20 is blocked. Therefore, since the diffracted light of the reflected light from the surface of the wafer 20 (alignment mark forming surface) is displaced with respect to the optical path of the specularly reflected light depending on the spatial frequency, the reflected light that reaches the spatial filter 38 allows only diffracted light (including scattered light) to pass through.

The reflected light from the wafer 20 that has passed through the spatial filter 38 is focused by three condensing lenses, and then irradiated onto the photodetector 50.

On the other hand, the polarization plane is rotated 45 degrees by the 45 degree rotator 32, and the fringe beam of the S polarization component, which is divided into a P polarization component and an S polarization component by the PBS 33, is reflected by the PB 833 and enters the PBS 40. . The number of fringes incident on the PB 840 is reflected and turned into circularly polarized light by the λ/4 plate 41, and then reflected by the mirror 42 and is perpendicularly incident on the alignment mark forming surface of the reticle 21. Then, a beam profile of a fringe having an interval PO as shown in FIG. 4 is imaged on the reticle 21 irradiated with the fringe beam. Reticle 21
Similarly to the case where the beam profile imaged above is irradiated onto the wafer 20, its fringes flow and scan the alignment mark in the shape of a diffraction grating formed on the reticle 21. When the alignment mark is scanned, reflected light such as specularly reflected light and diffracted light is generated with respect to the irradiation beam. The reflected light containing the positional information of the reticle 21 is reflected by the mirror 42, and most of it is converted into P-polarized light by the λ/4 plate 41 and transmitted through the PBS 40. The reflected light from the reticle 21 that has passed through the PBS 40 passes through the lens 43 and reaches the spatial filter 44 . This spatial filter 44 is provided with one for each reticle 21 surface by a lens 43, and forms a Fourier surface. Therefore, reticle 21
Of the reflected light from
only scattered light (including scattered light) passes through. The reflected light from the reticle 21 that has passed through the spatial filter 44 is reflected by the condensing lens 45.
After the light is focused, it is irradiated onto the photodetector 51.

When reflected light from each beam irradiation system for the same fringe beam is incident on the photodetectors 50 and 51, photoelectric signals φ1 and φ2 are outputted according to the intensity of the reflected light. The output photoelectric signals φ1φ2 are both input to the signal processing circuit 52.

Here, the photoelectric signal φn photoelectrically converted by the photodetector 50, 51 becomes an alternating current signal whose frequency is equal to the bit frequency as shown in FIG. The amount can be expressed by the phase difference θ between the photoelectric signal φ1 and the photoelectric signal φ2.

Next, a procedure for aligning the wafer 20 and the reticle 21 will be described. The wafer 20 is roughly positioned using a pre-alignment device (not shown).

0 After the rough positioning by the pre-alignment device is completed, the wafer 20 is then placed on the wafer stage. Thereafter, the chip to be exposed is moved to the calculated center coordinates of the reticle 21. In this state, the photoelectric signal φ1 obtained from the wafer beam irradiation system and the photoelectric signal φ2 obtained from the reticle beam irradiation system have waveforms having a phase difference θ as shown in FIG. And the signal processing circuit 5
2, the phase difference between the photoelectric signal φ1 and the photoelectric signal φ2 is detected, and the amount of relative positional deviation between the wafer 20 and the reticle 21 is calculated by calculating the amount of positional deviation−LOXθ/360. For example, the interval LO between the mark rows of alignment marks formed on the wafer 20 and the interval PO between the fringes of the fringe beam are 8 μm, and the phase detection accuracy is 0.
．． If the angle is 5 degrees, the position detection accuracy for 20 wafers is 0.0.
It becomes 1 μm.

In this way, the amount of relative positional deviation between the wafer 20 and the reticle 21 in the uniaxial direction is detected.

Similarly, the wafer 20 is precisely aligned based on the amount of positional deviation 1 in the axial direction perpendicular to the axial direction 1, and the position t1 determined in the two axial directions perpendicular to the axial direction. After the eyelid alignment is completed, the reticle is placed above the reticle 21, and the shutter in the illumination optical system is released for a predetermined period of time.
A projected image of a circuit pattern or the like formed on the reticle 21 is transferred to a predetermined printing position on the wafer 20.

According to this embodiment, the P-polarized light component and the S-polarized light component of the same fringe beam are irradiated onto the alignment marks of the wafer 20 and the C reticle 21, respectively, and the wafer is determined based on the phase difference of the reflected light. Since the position information of the wafer 20 and the reticle 21 is detected, the configuration of the optical system can be simplified, and the accuracy of the relative position information between the wafer 20 and the reticle 21 can be improved. can. In addition, the deviation amount of the positional deviation obtained by setting and averaging n mark rows and n fringe numbers is calculated using a general statistical method.σn=σ/~r]−(However, when σ is one When the number of fringes and the number of mark rows of alignment marks 2 are five as in this embodiment, the deviation amount σ5 of the positional deviation amount is expressed as σ5=σ/Jco. Since the amount of positional deviation can be detected with an extremely small error, the accuracy of position detection can be improved. Furthermore, since the amount of positional deviation is detected using the phase difference of the alternating current signals, the position detection accuracy can be further improved by taking a temporal average. Note that accuracy can be further improved by increasing the number of mark rows and fringes. Also, since the positional information of the wafer 20 and reticle 21 is read by the fringe beam flowing at the heat frequency, for example, if the heat frequency of the fringe beam is set to IMHz and 1000 pieces of phase difference information are captured at the sampling frequency of IMIIz, , positional deviation can be detected in 1/1.000 seconds, dramatically reducing the time required to detect positional information compared to obtaining wafer positional information by moving the wafer stage. can. In addition, since the relative positional deviation between the wafer 20 and the reticle 21 is measured sequentially3, if a fine movement mechanism with a resolution lower than the wafer stage length measurement resolution is used, it is possible to obtain either the measurement resolution by this method or the fine movement resolution, whichever is higher than the resolution. It is possible to perform rough alignment.

In the above embodiment, an array of five diffraction grating-like mark rows AMI to AM5 was used as the alignment mark, but depending on the configuration of the alignment optical system, a single grating pattern may be used. It may also be a simple line pattern.

When using a line pattern, it is desirable to detect scattered light generated at the stepped edges of the line pattern.

Furthermore, although the number of mark rows of alignment marks formed on the wafer 20 and the number of mark rows of alignment marks formed on the reticle 21 are the same, they may be different numbers of mark rows.

Further, in the above embodiment, the wafer 20 and the reticle 21
The spacing between the mark rows of the alignment marks formed in
If the distance between the wafer 20 side and the reticle 2]- side is relatively different from each other, it is not necessary to match them.

Furthermore, although the 45-degree Rochitaro 4 and the PBS65 are combined to cause interference between orthogonally polarized laser beams, it is also possible to use a polarizer to interfere with the laser beams, or use a polarizer to interfere with either the P-polarized component or the S-polarized component. The plane of polarization may be rotated by 90 degrees using a 90 degree rotator or the like to cause interference.

Also, a beam splitter 66, an afocal magnifying lens 67, photodetectors 68 and 69. Signal processing circuit 70
Since this is for maintaining the fringe spacing of the fringe beam at a predetermined interval, it may be removed from the apparatus after adjusting the frequency of the high frequency power source 71.

Further, in the above embodiment, the spatial filter 38.44, the condenser lens 39.45, and the photodetector 50.51 are used to detect the positional information of the wafer 20 and the reticle 21, but the spatial filter 38.44 Direct photodetector 5 at the position of
0.51 may be installed to detect components of scattered light or diffracted light (diffraction light of first order or higher order). Furthermore, the position of the spatial filter 5 3844 does not need to be on the exact Fourier plane with respect to the wafer surface and the reticle surface; for example, in the case of diffracted light, the 0th-order light component and the 1st-order or higher order light components are separated. As long as it can be detected, it may be defocused from the Fourier plane within that range.

〔Effect of the invention〕

As described in detail above, according to the present invention, the fringe light generated by adding i11 color light with a predetermined optical path difference and frequency difference is irradiated onto the alignment target, and the phase difference of the reflected light including the position information component is detected. Since the amount of positional deviation between the two objects to be aligned is determined, the configuration of the alignment optical system can be simplified.

In addition, even if the intensity of the light source or the reflection efficiency of the alignment mark changes, it can be measured without affecting the detection accuracy, and the amount of positional deviation of the alignment target can be accurately detected, allowing for highly accurate alignment. can be done.

Furthermore, the alignment time can be dramatically shortened.

6

[Brief explanation of drawings]

1 to 6 are diagrams showing an embodiment of the present invention. FIG. 1 is a configuration diagram of a beam irradiation system and a signal detection system, FIG. 2 is a plan view of alignment marks, and FIG. 3 is a diagram showing an example of the present invention. FIG. 4 is a diagram showing the configuration of the fringe beam generation system, FIG. 4 is a diagram showing the beam profile, FIG. 5 is a diagram showing the surface shape of the alignment mark, and FIG. 6 is a waveform diagram of the photoelectric signal. Figures 7 to 9 are diagrams showing conventional examples. Figure 7 is a configuration diagram of the alignment device, Figure 8 is a plan view of the alignment mark, and Figure 9 is the output intensity of the photoelectric signal and the scanning position. FIG. 20...6 Eha, 21...Reticle, 32.64.
...45 degree rotator, 33, 40.65... Polarizing beam splitter - 34, 37, 39, 43 degrees 45...
Lens, 34.41...λ/4 plate, 36°42...
Mirror 38.44... Spatial filter 50.51...
- Photodetector, 52... Alignment signal processing circuit, 6
1... Light source, 62... Acousto-optic element, 63.67.
... Afocal lens, 66... Beam splitter, 68, 69... 1st. Second light 7 detector, 70...signal processing circuit,

Claims (2)

[Claims] (1) A light source that generates monochromatic light; a fringe light generating means that synthesizes the monochromatic light output from the light source with a predetermined optical path difference and frequency difference to generate fringe light that flows at a heat frequency; fringe light irradiation means for irradiating the fringe light generated by the light generation means onto alignment marks formed on two alignment targets, respectively;
The fringe light irradiation means irradiates each alignment mark with the fringe light, thereby detecting reflected light including positional information components such as scattered light and diffracted light reflected from each alignment mark. a heat signal generated from the reflected light detected by the first light detection means and a heat signal generated from the reflected light detected by the second light detection means; An alignment apparatus comprising: a signal processing means for determining a relative displacement amount of the two alignment objects from a phase difference. (2) The alignment apparatus according to claim 1, wherein an acousto-optic element is used as the fringe light generating means.