JPH0269602A - Aligning device - Google Patents

Abstract

PURPOSE:To carry out aligning a first object with a second object with high accuracy by detecting gravity of the light quantity of reference luminous flux based on an alignment signal obtained from a first detector and zero-order diffracted light from a second alignment mark by a second detector. CONSTITUTION:First and second alignment marks 4 and 3 having the optical property are provided on each mask 1 and wafer 2 and aligning is carried out in the following manner by utilizing both luminous fluxs of signal luminous flux via each mark and the zero-order reference luminous flux from the second alignment mark 3. Namely, even if the gravity position of alignment light is changed by an inclination of the wafer 2 surface or an inclination of uneven application of a resist, a warp generated during the exposure process, etc., the relative gravity position between a reference signal and the alignment signal is detected, by which a deviation in the position is detected accurately without being influenced by the inclination of the wafer 2 surface. In addition, even if the gravity positions on optical sensors 11 and 12 of the alignment signal are changed, the deviation in the position between the mask 1 and the wafer 2 can be accurately detected by detecting the relative gravity position between the reference signal 8 and the alignment signal 7.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Field of Application) Undeveloped II relates to alignment devices.
Relative positioning of the mask and wafer when exposing and transferring a fine electronic circuit pattern formed on a first object surface such as a wafer onto a second object surface such as a wafer.
The present invention relates to a positioning device suitable for performing alignment.

(Prior Art) Conventionally, in exposure apparatuses for semiconductor manufacturing, relative positioning of a mask and a wafer has been an essential element for improving performance. Particularly in alignment in recent exposure apparatuses, alignment accuracy of 1, for example, submicron or less is required in order to increase the degree of integration of semiconductor devices.

In many alignment devices 2, so-called alignment patterns for alignment are provided on the mask and the wafer surface, and positional information obtained from these patterns is used to align both. As an alignment method at this time, for example, detecting the amount of deviation between both alignment patterns by performing image processing, or
As proposed in No. 4037969 and Japanese Unexamined Patent Publication No. 157033/1983, a zone plate is used as an alignment pattern, and a light beam is irradiated onto the zone plate, and at this time, the light beam emitted from the zone plate is focused in a predetermined plane. This is done by detecting the

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

FIG. 7 is a schematic diagram of the positioning device of Yoto using zone plates.

The f1r light flux emitted from the light source 72 at the time I''/l passes through the half mirror 74 by 1lll, and is condensed by the condensing lens 76 onto the condensing beam 1jI, 78, and then the mask alignment pattern 68a of the mask 68 and the support stand. 62, wafer alignment pattern 6 of wafer 60 side 1-
Irradiate Qa. These alignment patterns 68a
, 60a is composed of a reflective zone plate, and each number/1! Light Ij! , 78, a focal point is formed on the t-plane 1- which is orthogonal to the optical axis. At this time, the deviation of the condensing position t of the f-plane I- is detected by the condensing lens 76 and the lens 80 on the detection surface 82.
Light is guided to L for detection.

Based on the output signal t) from the detector 82, the control circuit 84 drives the drive circuit 64 to position the mask 68 relative to the wafer 60.

FIG. 8 is an explanatory diagram showing the imaging relationship of the light beams from the mask alignment pattern 68a and the wafer alignment pattern 60a shown in No. 71m.

At f'A, a part of the light beam diverged from the focal point 78 is diffracted by the mask alignment pattern 68a, and the light beam is focused near the focal point 78 indicating the mask position +:j7
8 Form a. In addition, some of the other luminous flux is passed through the mask 6.
8 as 0th-order transmitted light and transfers it to the wafer 6 without changing the wavefront.
The light is incident on the wafer alignment pattern 60a on the 0th plane. At this time, the light beam is diffracted by the wafer alignment pattern 60a, passes through the +1g mask 68 as zero-order transmitted light, and is focused near +j, j7a. Form. In the figure, the light beam diffracted by the wafer 60 or the condensed light f
When forming the mask 68, the mask 68 has a wide end shape.
Acts as E.

The position of the condensing point 78b by the wafer alignment pattern 60a formed in this way is set on a plane perpendicular to the optical axis including the condensing point 78, depending on the deviation Δσ of the wafer 60 with respect to the mask 68. Δσ° is formed as a fitting deviation corresponding to the deviation Δσ.

In this method, if the wafer surface is tilted with respect to the mask surface, the mask holder surface in the conductor exposure equipment, etc., and the ground plane with the exposure equipment, the light flux incident on the sensor beam will be reduced. The center of gravity position changes and alignment 5.
S yells.

In general, providing an absolute coordinate system for the sensor 1] and setting its reference thickness I' is another alignment,! '1 Difference factor 1 For example, inclination of the wafer surface due to warpage or deflection, variation in the center of gravity position of the luminous flux due to uneven coating of resist, variation in the east emission angle of the oscillation wavelength 1 oscillation output light of the alignment light source, variation in sensor characteristics, and Due to fluctuations caused by repeated alignment punt positions, it becomes very difficult to set the origin at a high degree of precision.

(Problem to be solved by the invention) Unexploded+1 allows for easily setting a reference point with high precision when aligning a first object such as a mask and a second object such as a wafer. Particularly, in the embodiment, which is characterized by providing an alignment device with rotation for precise alignment, the 0th order diffracted light from the alignment mark on the wafer surface is used as a standard light beam (reference light beam), and the wafer of the reference light beam is used. The effect of the movement of the center of gravity on the sensor with respect to the inclination of the surface is equal to the alignment light flux (signal light flux), and the change in the order of the alignment head is also equal to the reference light flux or alignment light flux, so that the reference light flux and This position makes it possible to equivalently cancel out error factors such as fluctuations in the relative position of the alignment light beam on the sensor, tilt of the wafer surface, and fluctuations in the position of the alignment head, making highly accurate alignment possible. Matching is intended for this offer.

(Means for solving the problem) A first optical element having a function as a physical optical element on the first object plane
forming an alignment mark, forming a second alignment mark having a function as a physical optical element on a second object plane, and directing diffracted light generated when a beam of light is incident on the PI4i alignment mark to the second alignment mark; The light was incident and diffracted by the first and second alignment marks.)
The light of the luminous flux: 11 ton centers are detected by the first detection stage F, and the alignment t'J (taken from the first detection stage) and the second
Light of the reference beam based on the 0th-order diffraction likelihood from the alignment mark 11) Detect the center of change using the second detection stage F, and from the second detection stage? 'Using both signals of the reference signal that is
The first glue body and the second object have been positioned.

(Example) 1st Z is a schematic diagram of the main part of the first example of the unreleased IJ1. In IA, l is the first object, for example, a square. 2
is a second object, for example a wafer, which is aligned with mask l. Reference numerals 4 and 3 denote first and second alignment marks, respectively, which are provided on the first surface I- of each mask and the second surface 1-1 of the wafer. Reference numeral 5 denotes a first iVj illumination mark, which is provided adjacent to the first alignment mark 4 on the first surface of the mask. 1st. 2nd alignment mark 4.3 and 11th
The 7-light mark 5 is made of a crater lens such as a Fresnel zone plate, and the mask 1 side F and the wafer 2
Blood 1. It is located at Skreitz Line 9.101-. 7 is the alignment luminous flux, I! II signal beam 8
These light beams 7 and 8 are emitted from a light source in an alignment head (not shown) and are collimated to a beam diameter of 11. 100 is a CPU, 101 is a mask drive mechanism, 102 is a sea urchin stage drive mechanism, 1
03 is a wafer stage.

In this embodiment, the types of light sources include semiconductor lasers, H, -N, lasers, A, lasers, and other light sources that emit coherent light beams, and light sources that emit non-coherent light beams, such as light emitting diodes.611. The second sensor (photoreceiver) serves as a first detection stage and a second detection stage, respectively, and is made up of, for example, a CCD or the like that receives the alignment light beam 7 and the reference light beam 8.

In this embodiment, the alignment light beam 7 and the reference light beam 8 each have 17
After being incident on the alignment mark 4 and the first reference mark 5 at a predetermined angle, the beam is transmitted and diffracted. Of these, the alignment light beam 7 is reflected and diffracted by the second alignment mark 3 of the wafer 2J-, and is incident on the surface of the sensor 11.

On the other hand, the reference light beam 8 is reflected across the width of the second alignment mark 3, that is, undergoes zero-order reflection, returns along the return optical path, enters the first reference mark 5, is diffracted once, and enters the second surface of the sensor. .

and sensor 11. Second, the alignment light flux and the reference light flux incident on the sensor surface H are detected at position t6, and the sensor 11. The mask l and wafer 2 are moved in the direction perpendicular to the scribe line 910 (y direction) using the output signal from the second
The position is being aligned.

Here, the light v-centering of a luminous flux is defined as the integral value or O vector when integrated over the entire cross section [nl] by multiplying the position vector of each point in the cross section by the light output of that point. It is also possible to take the order n of the point where the light intensity peaks as one representative point.

Next, the first and second alignment marks 4 in this embodiment
, 3 and the 11th jjlQ mark 5 will be explained.

Alignment marks 3 and 4 and reference mark 5 are 3 //
Fresnel zone plates with different values of focal length (
or a crating lens).

)11 alignment mark 4 and :jSts alignment mark 5 =j method is 140gm in each scribe line direction and 50m in the scribeline width direction (y direction).

The dimensions of the second alignment mark 3 are 280 gm in the scribe line direction and 50 gm in the scribe line width direction.

In this embodiment, the alignment light beam 7 and the reference light beam 8 both have an incident angle of 10" with respect to the mask 1, and
The projected component onto the surface is incident perpendicularly to the scribe line direction (x direction).

The alignment light beams 7 and 6 that entered the mask l at these predetermined angles! ! ,! Koto's 8 luminous flux/l becomes a converging (or emitting) light as a result of the Cratein Darens 45, and its 1st likelihood line from 7th mask 1 forms a predetermined angle to the normal line of mask 1. It's ejecting like that.

And the first alignment mark 4 and the 1911j+7-
The alignment beam 7 and the reference beam 8 which have been transmitted and diffracted through the wafer surface 2 are 1 each (2380gm and 20.11gm in the vertical direction of the wafer surface 2
The light is focused on the point 07yL.

At this time, the alignment mark 4 and the first Jjjj (+ mark 5's b, j', , l, and ') are each '/268IL.
m, 20.137 μm. Also, mask l and wafer 2
The barrier between the two is set to 30 μm.

The light transmitted through alignment mark 4 and diffracted is wafer 2 +
The second alignment mark 3 of r+i l: receives the gate (convex) lens action, and the sensor 11 surface 1 serves as the first detection r stage.
The light is focused on the - point of -. At this time, sensor 11 surface 1.
The light enters with a misalignment of the alignment mark 43 and a misalignment of the input device of 1-, which magnifies a sunken star. As a result, the sensor 11 detects a change in the K center (Q position) of the incident light beam.

In this embodiment, when the positional deviation between the mask 1 and the wafer 2 is 0, that is, the alignment mark 4 of the mask 1F and the wafer 2
When the alignment mark 3 or coaxial thread is formed, the exit angle of the L ray of the alignment light beam from the wafer 2 is 51
1', and a sensor in which the projected component of the emitted light at this time onto the second surface of the wafer is perpendicular to the scribe line width direction (y direction), and is disposed at a predetermined device, for example, at a height of 20 mm from the second surface of the wafer. It is set so that the light is focused on 11 sides.

At this time, the mask l and the first . The power distribution of the lens action of the second alignment mark 4.3 determines the magnification magnification -V for the positional deviation.

Also, in the 111th (one of the light beams transmitted and diffracted by mark 5)
゛Ray of light or two wafers! In this embodiment, the component that is directly incident on - and reflected at the 0th order is the first reference jj of -
(It is incident on the + mark 5 at double right angle. Of the light beams of each order that are diffracted and transmitted through the first reference mark 5, the 0th order light beam is emitted in the normal direction of the mask 1 surface, and the 1st order diffracted light 1
The three rays emit the same optical path as the incident light flux in the opposite direction, resulting in second-order diffracted light! To! Six points are set so that the light beam 8 is incident on one point on the second surface of the sensor. At this time, the reference light beam 8 and the signal light beam 7 are separated at the sensor surface by the difference in their respective emission angles with respect to the normal to the mask surface.

As described above, in this embodiment, the capacitive light beam 8 is reflected across the width of the two wafer surfaces, and the O-order reflected light beam, which is not subjected to lens action, is used to measure the error of the positional shift meter 814 due to wafer surface tilt, pink-up fluctuation, etc. ing.

Incidentally, the exit angle of the F ray of the reference beam 8 from the mask 1 surface is 8
The projected component is set to be orthogonal to the width direction of the scribe line.

At this time, the reference light beam passing through the first reference mark 5 is transferred to the second sensor even if there is a change in positional deviation between the mask l and the wafer 2.
The centering position of the incident light on the surface is always constant.

In this way, in the unexamined example, the mask surface l''.

The grade ink lens indicated by the 19th reference mark 5 has a relatively long focal length in which the reference iμ light flux passes through one surface of the mask, reflects off the second surface of the wafer, and reaches the second surface of the sensor.

In this embodiment, the reference light beam 8 is a light beam that is reflected off the two wafer surfaces and then diffracted in the 12th order at the first reference mark 5 on the first mask surface, and the signal light beam 7 is the light beam that is diffracted by the second alignment mark 3 on the second wafer surface. Although a light beam that is reflected and diffracted in the -1st order is used, it is also possible to use a +2nd-order diffracted light and a +1st-order diffracted light to emit the light to the east of the incident light.

Furthermore, the first diffraction order is not necessarily limited to the second order and the first order, and other orders (other than the 0th order) may be used.

In this case, the magnification sensitivity of the positional deviation of the signal light flux changes accordingly.

The basic algorithm for alignment (lateral shift detection and control) of the mask 1 and wafer 2 according to this embodiment is as follows.

! O First, after measuring the light intensity distribution of the positional deviation signal beam on the sensor 38-2, the previously defined light intensity gravity center position X is determined.

(2) At this time, the light I of the reference luminous flux of the sensor 3B-21-
From J minutes/+j, calculate the light intensity and center of gravity position X and l of the rainbow luminous flux.

・Determine 3・X and
Find Lf'1.

(4) Relative positional shift of mask l or wafer 2;
The wafer stage 103 is moved by the wafer stage drive mechanism 102 in order to relatively move the wafer stage,
F.

・5) Perform middle to 3J operations to remove mask 1 and wafer 2.
) Determine whether the relative positional deviation hardness Δf2 of iil is within the allowable range of i1.

(6)Δ5. or permissible ((until it falls within the range of (tortoise) ~ 1
Repeat 5+.

FIG. 9 shows a flowchart of the #1 main point of rX continuation 1-.

Note that the mask driving mechanism 101 may be used to move the mask in the positional deviation compensation IF.

Next, the first and second alignment marks 4 in this embodiment
An example of a method for manufacturing the first reference mark 5 (grating lens) and the first reference mark 5 will be described below.

First, the mask marks 4 and 5 are designed so that a predetermined ascending beam of light enters at a predetermined angle and is condensed at a predetermined position. In general, the pattern of a grating lens is an i-fringe pattern on the lens surface when a floating light source is placed at the light source (object point) and image point, respectively. Now, mask l +l'j l- as in No. 1171
Define the coordinate system of Here is Hara 1. - is located at the center of the scribe line radius, and has two axes: an X-axis in the direction of the scribe line force, a y-axis in the width direction, and a normal force direction to the mask surface l. It is incident at an angle α attached to the normal line of the mask surface l, and its projected component, or the ascending light beam orthogonal to the scribe line nozzle j direction, passes through the mask mark and is diffracted, ! &! Light spot (X+, y+, Z+)
The equation of the curve group of the grating lens that images the image at 1/ of
-P2m1 person/2-...-(+)pl(x)=X-X
+”y-y+2”Z+”Pt”X+
``y+'' can be calculated by ``Z+''. Here, input is the appropriate wavelength for alignment, and m is an integer.

The chief ray is emitted from the person at an angle α, and the mask surface 11. Assuming that the ray passes through the origin +7° of XYl, Z+) represents the difference in the optical path length of the rays reaching the point (XYl, Z+), and the right side shows that this difference is m/2 times the wavelength longer (shorter) than the principal ray due to the ratio of m. Figure 2 (A) shows the first part on the mask l.
Alignment mark 4, first reference mark 5 in the same figure (B)
shows.

- On the other hand, the grating lens 3 on the wafer 2 is a single point light source designed to condense the spherical wave emitted from a predetermined point light source onto a predetermined position (sensor surface F). And let g be the gap during exposure of wafer 2, then (X+
, Y+, Z+ g). (y is a variable) Mask 1 and wafer 2 are aligned in the y-axis force direction, and when alignment is completed, +, 'i (
If the alignment light is to be focused at the positions (・・・・・・・・・・・・
...It is expressed as (2).

Equation (2) shows that the wafer surface is at z=-g, the Kei ray is on the mask plane L, the original I and I on the wafer surface are: $(0,0-g),
Furthermore, assuming that the ray passes through j, (X2.y22+) on the sensor surface, the grating (x, y22+) on the wafer surface I-
, -g) and the difference in optical path length between one ray and the other is an equation that does not satisfy the condition that the difference in optical path length is an integral multiple of a half wavelength.

FIG. 2(C) shows the second alignment mark of the wafer 2)4.

Generally, a zone plate (grating lens) for a mask is formed at the intersection 11 of two areas: an area through which light rays pass (transparent area) and an area through which light rays do not pass (shading area).
, 1 is formed as an amplitude-type cladein/' Jc (1. Also, the zone plate of the wafer river is formed, for example, as a 10 phase r pattern with a rectangular cross section. The phase case r- is formed as, for example, the same village. This is the concavo-convex pattern of Ne1.(1)
, In Equation (2), the outline of the grating is defined as IN/17I, which is an integer multiple of the wavelength of 1, for a ray of light. Mask l! This means that the line width ratio of the transparent part and the light-shielding part is lI for the grading lens of 2, and that the ratio of the line and space of the rectangular case r- is 1+l for the grating lens of wafer 2I.

Mask 11. The grating lens was formed by rolling a grading lens pattern of a reticle formed by f/EM EB exposure on an organic thin IQ +- made of polyimide.

Also, 1 of wafer l. The marks were formed by forming the exposure pattern of the wafer on the master 1- and then transferring it by exposure.

Next, a description will be given of the relationship between the signal light, which is the alignment light, and the reference light, which are incident on the sensor (for example, one-dimensional accumulation type one-dimensional cco''9) as the detection means in this embodiment.

In an unimplemented example, 9 jpl, the light and alignment angle (, +) are aligned with the normal to the wafer surface at an angle of q86.!5°, and the wafer surface projection with respect to the flyline direction. The components are emitted perpendicularly to each other.Sensor 11.Second
Spatial arrangement +: t,'f The beam is tinted so that it is incident at approximately the center position of the sensor when the alignment is completed.

Sensor 11. The second center distance is 2mm and approximately 0.1g
The rotation of Si at mj/i degrees is given as follows:
The plate has its normal line, alignment light output angle and valley light output (
It is arranged substantially in alignment with the bisector of No. 11.

Sensor 11. The second size is 1) The sensor 11 for light is 1 mm wide and 6 mm long, and the second sensor for valley light is 1 mm wide.
mm, and the length is 1 mm. Also, the size of each pixel is 25
End mX500. Lm-(Yes.

Each sensor measures the center position in the incident light, and the output of the sensor is the total light in the receiving area;
It is processed. As a result, even if the output of the alignment light source changes slightly, the measurement output from the sensor system ((I is IF
The center position of the ball is precisely set. The resolution of the centering position of the sensor may depend on the power of the alignment light, for example, at 50 mW and a wavelength of 0.83 μm, and as a result of measurement using a conductor laser, it was 0.2 gm.

A design example of a grating lens for a mask and a grating lens for a sea urchin according to this embodiment is shown in FIG. Move the teton/shikuq position. Therefore,
If there is a 0.01 μm (Q) misalignment between the mask and the sea urchin, an effective centering movement of 1 gm will occur on the sensor surface!, and the sensor system will measure this with a resolution of 0.2 gm. I'll come.

In this embodiment, C wafer surface 2 is within yzll'+i
If rad#iA1, sensor 2- is 5)
Koei causes the center of gravity to shift by approximately 40 μm. On the other hand, reference IV (Kouei 8 is also axially symmetrical with Koei 7, and the second sensor passes through an optical path with the same optical path length, causing the same center of gravity to shift as the signal light. This causes the sensor In other words, if signal processing is performed to output the difference in the effective center of gravity position from each q sensor, the output signal from the sensor system will not change even if the wafer surface is tilted in the yz plane. .

force, when the wafer is tilted in the xz plane, the signal honors 2,3
Whether the center of gravity of both the igl light flux shifts in the width direction perpendicular to the length direction of the sensor, or this can be detected on the sensor (
Since the direction is perpendicular to the direction of the movement of the center of gravity due to the Q misalignment, there will be no effective alignment error even if there is no reference light.

Furthermore, if the position of the alignment head containing the alignment light source, projection lens system, sensor, etc. is changed relative to the mask-wafer system, for example, suppose the head is moved in the 54 my direction relative to the mask. . At this time, the signal light causes an effective centering shift of 5 .mu.m at the sensor 11, ); whereas the reference light also causes an identical center of gravity shift of 5 gm at the second sensor.

Similarly, lOpLm in the z direction between the mask surface and the head.
When a fluctuation occurs, two signals are sent to the light source sensor 11 and ! The second sensor for consolation causes a center of gravity shift of 10 μm.

Therefore, the final output from the sensor system is
) The difference signal between the barycenter position output of the light and the output of the reference light at the center of gravity does not fluctuate by 16 seconds.

Also, see IK for the variation of 佼t in the X width direction! It can be seen that even if there is no error, there is no essential alignment error.

It should be noted that the alignment mark 3.4 and the reference mark 5 do not need to be set to have a distance of 1', -1, (1) which is different from each other as in this embodiment. You can change the distance to focus.

As shown in this example, in order to cause the signal light and the reference light to move in exactly the same direction on the sensor depending on the fluctuation factors such as the tilt of the wafer surface and the change in the position of the alignment head, the mask is required. It is an essential condition that the optical path lengths of the signal light beam and the reference light beam until they reach the sensor l2 are substantially equal when the relative positional deviation between the two and the sea urchins is around 0, and this implementation is a structural requirement for that purpose. In the example, 1) As a physical optical element f having a predetermined lens action, two types of focal points ljj are placed on the mask surface l-, one for signal light, one for valley, and one for light flux.
On the wafer surface, a signal light Higashikawa grating lens with one type of focal length and approximately the same size as the alignment mark formed on the mask surface is formed. In other words, either one of the alignment marks is subjected to the lens effect, or neither of the alignment marks is subjected to the lens effect.

蒐2) The reference light receiving sensor and signal light receiving sensor are the same.
It is formed on the substrate surface 1-, so that the normal line of the substrate surface is approximately in the line of the bisector of the signal beam emission angle from the wafer surface and the reference beam emission angle when the positional deviation is 1 position, 1 position is 0, and the reference beam emission angle is 0. Placed.

(The grating lens pattern was designed so that the wafer surface was orthogonal to the two planes formed by the two optical paths emitted from the wafer surface when 3+positional deviation was 0.

The above is a must! Regarding (ri) above, rather than ff, this is an example of the necessary configuration requirements for sensor arrangement that causes equal center of gravity movement.Another example is that when the position is shifted by 1 position, each sensor is placed on each sensor surface. The sensors may be arranged so that the light enters directly and each optical path length from the wafer surface is equal.

Furthermore, as another configuration regarding l2, the grating lens pattern may be sunk so that the bisector of the angle formed by the two optical paths is orthogonal to the positional deviation detection direction.

Further, as shown in the flowchart of FIG. 9, the signal light and the reference light do not necessarily have to match the positional deviation detection direction components in the emission direction to the sensor surface.

For example, the positional deviation detection direction oblique component of the distance statement between the heart position of the signal light and the reference 1 consolation light conversion position is a predetermined value ε. The positional deviation measurement and the positional deviation sleeve IF control may be performed so that the predetermined allowable IIIQ range ε1≦(.≦(2) is satisfied.

Alternatively, the differential output value between the output value of the signal light receiving sensor and the output value of the reference light receiving sensor or the allowable value 1 centered around a predetermined value.
Similar measurement control may be performed to fit within the range.

The third UA (A) is a schematic diagram of main parts of the iR2 embodiment of the present invention.

In the same [A], elements that are the same as those shown in FIG. 1 are numbered.

In this embodiment, the displacement detection force direction is scribe line 9. There is a direction of lO force. At least one of the first alignment mark and the reference mark 5 is formed in a plurality of regions.

3Z (B) is a schematic diagram of the II alignment mark 4 and reference mark 5 on the mask 1 surface, and FIG. 3 (C) is the wafer 2
3 is a schematic diagram of a second alignment mark 3 on a surface; FIG.

As the reference light beam 8, as in the first embodiment of the first IA, O-order anti-Q (diffracted light) from the wafer 2 surface 1- is used.

In this embodiment, there is only one incident light beam 6, and this is separated by an alignment mark to form a signal light and u! To! I'm getting light.

Note that the focal lengths of the marks on the mask 1 and the wafer 2 are the same as in the first embodiment.

As shown in this example, after each of the reference light and the reference light is transmitted and diffracted through the masked physical optical element, the wafer surface is
As a result of configuring the physical optical element between the parts of - and area H to diffract the reflected light, the influence of the local tilt of the wafer surface 1- can be suppressed by the optical centering position 2114IJ- of the reference light beam. .

In this example, by setting only the Shini7 Mitsukawa mark on the surface of the sea urchin, the structure of the mark is simplified, unnecessary diffraction order light components are reduced, and positional deviation (f) signal light is It is possible to prevent the private signal light 11 from decreasing.

FIG. 4(A) is a schematic diagram of a main part of a third embodiment of the present invention.

In the same figure, is the key point and times/elements shown in m l I'K times? It is numbered 1.

Figure 4 (B
), the signal light Kretein lens pattern and the reference light Kretein lens pattern or Ol amplitude i
Mi! As an i-type pattern, 1iii pi- is formed in 2 tons.

4t71fC) is a schematic diagram of the second alignment pattern provided on the wafer 2 and 1-.

In order to be smaller than the same IA, only a grating lens pattern for the signal beam is set on the wafer 2 side 1:, and the 0th order reflected light from the wafer surface 2 is used as the reference beam for detecting the inclination of the wafer surface. The grating lens on the mask surface further allows the second-order diffracted light beam to be transmitted by the second reference light sensor within the alignment mark. The power arrangement of the signal light beam/l and the reference light grating lens is the same as in the first embodiment. This embodiment also provides the same effect as the 3171st embodiment.

FIG. 5 is a schematic IA of the main part of the fourth embodiment of the present invention.

In the same figure, the same element as the first tel ・ρ748
is numbered 4.

In this embodiment, the (+/positional displacement detection force direction) is the scribe line direction (X direction).

In addition, one example is the mask surface 1. Only the mark 4 for the alignment signal light is set for the mark 4, and only the mark 3 for the alignment signal light is set for the wafer surface l-. The masked mark and the wafer 1 mark were shifted in the cross direction by 20 μm in succession.

In this embodiment, the light beam transmitted through the 0th order mark of the mask surface 1- and reflected on the wafer surface 1 mark in the 0th order is used as the reference light beam, and a light receiving lens system and sensor for receiving the 1st reconnaissance light east are used. The second one was set. Also, the above-mentioned 1st to 3rd
As in the second example, in order to avoid the influence of undesirable diffracted light, a system was constructed in which the light enters and exits obliquely to the normal line of the mask surface.

The light is 1 in the yz plane with respect to the line.
After entering at 1', it receives lens action and becomes 1'y to the normal to the wafer surface.
Inject it onto the wafer surface at an angle of 2'' in the z blood, and then diffract the mark 3 for 1 position on the wafer surface as a primary diffraction, and make an angle of 16'' in the yz plane with respect to the normal to the wafer surface. We used light that would emit light.

In the XZ incident plane, the alignment light beam is
When completed, it is designed so that the light enters and exits directly from the mask surface or wafer surface at 1F. Other mark sizes and Kretein lens power arrangement 1 position displacement knee detection magnification -V are the same as in the first to third embodiments.

In the first to third embodiments, the reference light beam is zero-order reflected from the wafer surface, and the reference light beam is reflected from the mask surface 1. After being second-order diffracted at the reference mark, the light is focused on the sensor. In these valley embodiments, the reference light beam was configured to diffract the mark on the mask surface in one order of order. See! The optical path of the consolation beam is diffracted from the mark on the mask l + ni l: in the 1st order, and the mark on the wafer 2 surface 1 is reflected in the 0th order, and then the mask is
Next, using a beam that performs iAJ diffraction, the reference beam mark (focal length 11, 20.167 mm) provided on the mask surface] is used. The 1st-order diffracted transmitted light beam makes an angle of 5° with respect to the wafer surface normal in the yz plane cross section,
The surface mark is reflected to the sea urchin and is emitted from the mask surface at an angle III' of lk4 and 5'' to the normal line of the mask surface.

Note that the application of the present invention is not limited to the positioning mechanism of a conductor I2I device; for example, the positioning of a gap in a hologram element f-setting during exposure reproduction of a hologram, or the positioning of a gap in a multicolor printing machine. Widely applicable to position alignment, optical parts, alignment 1 interval measurement of optical measurement system adjustment, etc.! 1117.

Also, in the xz plane cross section, when alignment is complete, Figure 6 CB)
As shown in the figure, both the signal light 7 and the reference light 8 are projected so as to enter and exit in a direction perpendicular to the mask and wafer surfaces (normal direction).

The signal light is emitted from the wafer surface at an angle of 16@. Other aspects such as the exit angle of the alignment light beam, the power arrangement of the grating lens, and the mark size are the same as in the first embodiment.

(Effects of the Invention) According to the invention I, the first and second
Place each alignment mark in the first position. Second? For example, by using both the signal light flux passing through each mark and the reference light flux based on the zero-order reflected light from the second alignment mark of the second body,
When aligning the mask as the h-body and the wafer as the second object, the following effects can be obtained.

(b) Even if the t6 position of the alignment light changes due to an inclination of the wafer surface, uneven coating of the resist, or local inclination such as warping caused during the 'A-first light process, the reference signal light cannot be used. By detecting the centering position relative to the alignment signal light, 1
[It is possible to accurately detect fQ misalignment.

(b) Since the position of the alignment head has changed relative to the mask, even if the center position of the alignment signal light sensor has changed, the IK light and the alignment signal light can be used. Relative center position: By performing d detection, +E can be adjusted without being affected by the positional deviation of the
It is possible to accurately detect the (Q misalignment) between the mask and the wafer.

(c) Furthermore, even if the gap between the mask and the wafer changes, the relative center of the reference signal light and the alignment signal light changes even if the t center (! position) of the alignment detection direction of the signal light alignment sensor l- changes. By performing position detection, it is possible to accurately detect IF misalignment without being affected by gap fluctuations.

(d) It is possible to selectively receive the zero-order reflected light from the wafer surface even without providing a reference light mark on the mask surface L, thereby suppressing unnecessary diffracted light and eliminating local tilting of the wafer. It has a simple structure and is capable of producing excellent effects.

[Brief explanation of the drawing]

1st [A is a schematic IA of the optical system of the first embodiment of the present invention, 2nd f21 (A), [a), and (C) are a partial explanation of FIG. 1 IJI I;4, 3rd. 4゜Figures 5 and 6 show the second, third, fourth . Schematic diagram of the optical system of the fifth embodiment, 7th. FIG. 8 is an explanatory diagram of a positioning neck attachment using a conventional zone plate, and FIG. 9 is a flowchart of the first embodiment related to misfire 1!1. In the figure, 1 is the first object (mask), 2 is the second object (wafer), and 4.3 is the first object (wafer). 2nd alignment mark, 5 is first reference mark, 7 is alignment light, 8 is reference light, 9
, 10 is a scribe line, 11 is a first detection system (sensor), and 2 is a second detection system (sensor).

Claims (2)

[Claims] (1) A first alignment mark having a function as a physical optical element is formed on a first object plane, a second alignment mark having a function as a physical optical element is formed on a second object plane, and the first alignment mark is formed on a second object plane. The diffracted light generated when the light beam is incident on the alignment mark is made to enter the second alignment mark, and the center of light intensity of the signal light beam diffracted by the second alignment mark is detected by a first detection means, and the first detection means A second detection means detects the center of light intensity of a reference light beam based on the alignment signal obtained from the second alignment mark and the zero-order diffracted light from the second alignment mark, and the reference signal obtained from the second detection means is used. , a positioning device characterized in that the first object and the second object are positioned. (2) The reference light flux is incident on a reference mark that is provided adjacent to or overlaps with the first alignment mark on the first object surface and has a function as a physical optical element, and is diffracted at a predetermined order. 2. After that, the light beam enters the second alignment mark and undergoes zero-order reflection.
The alignment device described.