JP2003282392A - Alignment method and alignment device - Google Patents

Abstract

Abstract: PROBLEM TO BE SOLVED: To provide an alignment method and an alignment apparatus for realizing high-precision alignment in an exposure apparatus by avoiding erroneous detection of the position of an alignment mark without increasing the number of steps in a semiconductor device manufacturing process. I do. SOLUTION: The position of a _first alignment mark A1 formed of a step portion formed on an insulating film on a wafer 52 and having a metal film formed thereon is detected, and is formed on the insulating film on the wafer 52. a step of detecting the position of the alignment mark B ₁ consisting of the step portion which is not a metal film is formed thereon, the detection from the detection result of the position of the alignment mark B ₁ and the alignment marks a ₁ position of the alignment marks a ₁ and a step of obtaining an error, in consideration of detection error, and a step of correcting the position information of the alignment marks a ₁ to be used for alignment of the wafer 52.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of aligning a wafer or the like in an exposure step of a semiconductor device manufacturing process and an alignment apparatus.

[0002]

2. Description of the Related Art As a method of detecting an alignment mark in an exposure apparatus used in a semiconductor device manufacturing process or the like, a detection method based on an image processing system has become mainstream.
In the case of the image processing method, the alignment mark formed on each layer on the wafer is vertically irradiated with illumination light of a wide band wavelength. Then, the reflected light and the diffracted light from the alignment mark are condensed via the imaging optical system. An image of the alignment mark is formed by focusing the reflected light and the diffracted light which are collected on the image pickup surface of the image pickup device. The position of the alignment mark on the wafer is detected by performing signal processing on the image pickup signal thus obtained by the image pickup device.

The alignment mark detected as described above is composed of, for example, an array of step portions such as grooves formed in an oxide film or the like on the wafer. In the exposure step, a metal film is formed on the oxide film on which the alignment mark is formed, and then a resist film is formed on the metal film.
Then, the alignment mark composed of the step portion is detected to align the reticle for patterning the resist film. Next, the exposure film is irradiated with illumination light through the reticle to expose the resist film.

[0004]

An aluminum film to be a wiring layer, for example, is formed on the insulating film or the like having a step portion to be an alignment mark by a sputtering method or the like.
When the metal film is formed on the step portion in this manner, the original edge position of the step portion may be erroneously detected. The erroneous detection of the edge position will be described with reference to FIG. FIG. 7 is a sectional view of the alignment mark at the left and right positions of the wafer.

As shown in FIG. 7, on the wafer 100,
Insulating film 102 made of TEOS (TetraEthOxySilane)
Are formed. The insulating film 102 has a step portion 104 that forms an alignment mark. A metal film 106 made of an alloy of aluminum and copper is formed on the insulating film 102 on which the step portion 104 is formed. The metal film 106 is asymmetrically coated on the opposite edges of the step portion 104 formed in the insulating film 102.

As described above, the step portion 104 has the metal film 106.
Therefore, when the alignment mark position is detected in the exposure apparatus, the original edge position of the stepped portion 104 is erroneously detected.

Particularly, by the reflow sputtering method for enhancing the burying property of the metal film in the groove by the high temperature treatment for a long time,
When a metal film made of aluminum or the like is formed, erroneous detection of the original edge position of the step portion becomes significant.

Aluminum, which is one of the materials generally used as a metal film material, has a high light reflectance. Therefore, it has been difficult to construct an optical system that allows the position of the step portion formed in the insulating film existing under the aluminum film to be transmitted and detected.

Conventionally, in order to prevent the original position of the alignment mark from being erroneously detected as described above, the aluminum film formed on the alignment mark is generally removed. For this reason, one lithography process and one etching process using a dedicated reticle are required, and the number of semiconductor device manufacturing processes needs to be increased.

It is an object of the present invention to provide an alignment method and an alignment apparatus which avoid erroneous detection of the position of an alignment mark without increasing the number of steps in the manufacturing process of a semiconductor device and realize highly accurate alignment in an exposure apparatus. To do.

[0011]

The above object is to detect the position of a first alignment mark having a step portion formed on an insulating film on a wafer and having a metal film formed thereon, and to detect the position of the first alignment mark on the wafer. A step of detecting the position of the reference alignment mark formed by the step portion formed on the insulating film and not having the metal film formed thereon, and the detection result of the positions of the reference alignment mark and the first alignment mark. A step of obtaining a detection error of the position of the first alignment mark from the above, and a step of correcting the position information of the first alignment mark used for the alignment of the wafer in consideration of the detection error. It is achieved by the alignment method described below.

Further, the above object is to detect the positions of the first and second alignment marks having a step portion formed on the insulating film on the wafer and having a metal film formed thereon,
A step of detecting the position of a reference alignment mark formed on the insulating film on the wafer and having a step portion on which the metal film is not formed; the reference alignment mark and the first and second alignments. A step of obtaining a detection error of the positions of the first and second alignment marks from the detection result of the position of the mark, and a step of obtaining the detection error of the first and second alignment marks used for alignment of the wafer in consideration of the detection error. And a step of correcting position information.

Further, the above object is to detect a position of a first alignment mark having a step portion formed on an insulating film on a wafer and having a metal film formed thereon, and to detect the position of the insulating film on the wafer. Position detecting means for detecting the position of the reference alignment mark formed of a stepped portion formed on the surface of which the metal film is not formed, and the detection result of the positions of the reference alignment mark and the first alignment mark. A detection error measuring unit that obtains a detection error of the position of the first alignment mark and a position information correction unit that corrects the position information of the first alignment mark used for the alignment of the wafer in consideration of the detection error. It is achieved by an alignment device characterized by having.

Further, the above object is to detect the positions of the first and second alignment marks having a step portion formed on the insulating film on the wafer and having a metal film formed thereon,
Position detecting means for detecting the position of a reference alignment mark formed on the insulating film on the wafer and having a step portion on which the metal film is not formed, the reference alignment mark, and the first and second Detection error measuring means for measuring the detection error of the position of the first and second alignment marks from the detection result of the position of the alignment mark, and the first and the first and the second used for alignment of the wafer in consideration of the detection error. This is achieved by an alignment apparatus characterized in that it has position information correction means for correcting the position information of the second alignment mark.

[0015]

DETAILED DESCRIPTION OF THE INVENTION [Principle of the Invention]
We have analyzed the misdetection of the position due to the asymmetric coating of the alignment mark with the aluminum film.

1 (a) and 2 (b) are graphs showing wafer scaling (Wafer Scaling) and wafer rotation (Wafer Rotation) before and after the etching of the aluminum film formed on the alignment mark.

The alignment error of the current layer with respect to the lower layer can be classified into a linear error and a non-linear residual error. The linear error is divided into components of wafer expansion / contraction, rotation, translation, shot expansion / contraction, and rotation. The wafer that has undergone the process steps may actually expand and contract. The magnification component (coefficient) is corrected to the step pitch during exposure according to the expansion and contraction of the wafer.

When the alignment mark formed on the oxide film is covered with the aluminum film, the wafer magnification changes depending on the presence or absence of the aluminum film covering before and after the etching removal. Therefore, although the wafer does not actually expand or contract, it seems that the wafer apparently expands or contracts due to the asymmetry of the aluminum film.

A linear regression analysis can be used as an analysis method for the component analysis of the displacement within the wafer. The linear regression analysis of the displacement in the wafer is performed by measuring the displacement measurement device with Δxi, Δyi, the coordinates of the i-th displacement measurement mark in the wafer with Pxmi, Pymi, the wafer scaling with Mx, My, and the wafer rotation. Rx, Ry, position deviation measurement error and stepper random error (or stepper stage mix & m
the sum of the touch precision (including the random precision) is δxi,
Let δyi be the following formula

[0020]

[Equation 1] Therefore, the residual Σ (δxi) 2 and the residual Σ (δxi) 2 reflect the measurement accuracy and the random error in the stepper (or the step-to-stage mix & match accuracy (including the random accuracy)). Mix and match between stages in advance
Since the accuracy can be obtained, the measurement accuracy can be predicted. By performing the same analysis among the chips, the chip magnification deviation and the chip rotation can be calculated. We will separately add more detailed factors of aberration in the chip.

That is, the position of the alignment mark is measured, and linear regression analysis is performed for each linear component of translation (shift), expansion and contraction, and rotation. Since the stage of the exposure apparatus can move only linearly, the components are linearly matched. Since the accuracy of the stage of the exposure apparatus is extremely good, it is possible to consider all the components that do not have a linear relationship as measurement errors.

Note that ΔScaling_x_ in the graph of FIG.
1 μm and ΔScaling_y_1 μm are the differences before and after the etching removal of the aluminum film in the alignment mark having a width of 1 μm in the x direction and the y direction, respectively. ΔScaling_x_
2 μm and ΔScaling_y_2 μm are the differences before and after the etching removal of the aluminum film in the alignment mark having a width of 2 μm in the x direction and the y direction, respectively. Here, the difference is a deception of the position of the alignment mark due to the aluminum coating. For ΔRotation, CMP (C
Generally, it is considered that the position is rarely deceived by the rotation component, except for the case where the alignment mark has a habit in the rotation direction due to the rotation of the wafer in the chemical mechanical polishing).

As is clear from the graph shown in FIG. 1A, the misdetection of the position of the alignment mark has a great influence on the measurement result of the wafer magnification. On the other hand, FIG.
As is apparent from the graph shown in (b), the measurement result of the wafer rotation has very little influence or little influence.

However, the apparent wafer magnification is due to the asymmetry of the coating of the step portion with the aluminum film. Therefore, in the case of a sputtering process in which the stability and reproducibility of the coating are small, it is not constant between lots of wafers. In such a case, by optimizing the width of the alignment mark, it is possible to reduce the variation in wafer magnification from lot to lot. However, the wafer magnification cannot be constant between lots. In the first place, the optimization of the mark width cannot solve the erroneous detection itself of the position of the alignment mark.

In the present invention, the position for forming the alignment mark is devised, the metal film such as the aluminum film on the alignment mark is removed by a simple method, and the alignment mark and the metal film from which the metal film is removed are formed. By measuring the position coordinates with the alignment mark, erroneous detection of the position of the alignment mark is avoided and the alignment accuracy is improved.

First, in the present invention, the position for forming the alignment mark is formed, for example, in the vicinity of the outer peripheral portion of the wafer. The formation position of the alignment mark in the alignment method according to the present invention will be described with reference to FIG. FIG. 2A is a diagram for explaining the alignment mark formation position in the alignment method according to the present invention, and FIG. 2B is a diagram for explaining the alignment mark formation position in the conventional alignment method.

Conventionally, the alignment mark 10 is shown in FIG.
As shown in (b), it was formed for each shot of the wafer 12. Therefore, in order to prevent the original position of the alignment mark 10 from being erroneously detected, the aluminum film formed on the alignment mark 10 is generally removed. For this reason, one lithography process and one etching process using a dedicated reticle are required, and the number of semiconductor device manufacturing processes needs to be increased.

On the other hand, in the present invention, as shown in FIG. 2A, the alignment mark 14 is formed near the outer peripheral portion of the wafer 16. In this way, the alignment mark 1
By forming 4 in the vicinity of the outer peripheral portion, it is possible to easily remove the resist only in the vicinity of the outer periphery by an outer peripheral exposure apparatus that is less expensive than an exposure apparatus such as a stepper, and the aluminum film formed on the alignment mark. Can be removed by etching.

In the initial stage of process development, the aluminum film is removed by the above-mentioned simple method, and the alignment mark 1 formed of the step portion formed in the underlying oxide film is formed.
Expose 4 Then, the wafer magnification is measured in an ideal state where the aluminum film is removed, and the step of removing the aluminum film may be omitted as the process matures.

Further, the trend data of the measured wafer magnification is stored, whether the wafer magnification correction is necessary or not is judged from the accumulated trend data, and the wafer magnification is corrected if necessary.

As described above, the alignment mark is formed in the vicinity of the outer periphery of the wafer so that the aluminum film or the like formed on the alignment mark can be easily removed, and the wafer is formed in an ideal state with the aluminum film removed. Since the magnification is measured, the necessity of correcting the wafer magnification is determined based on the measured trend data of the wafer magnification, and the wafer magnification is corrected as necessary, it is possible to realize highly accurate alignment.

[Embodiment] An alignment method according to an embodiment of the present invention will be described with reference to FIGS. 3 is a schematic diagram showing the configuration of the exposure apparatus, FIG. 4 is a diagram for explaining a method for measuring the wafer magnification in the alignment method according to the present embodiment, and FIG. 5 is a trend chart of the wafer magnification measured by the alignment method according to the present embodiment. 6 is a graph showing an example, and FIG. 6 is a schematic diagram showing an apparatus configuration of an exposure apparatus when the alignment method according to the present embodiment is performed.

First, the exposure apparatus used in the alignment method according to the present embodiment will be explained with reference to FIG.

The exposure apparatus includes an alignment mark 30 formed on a wafer 28 placed on a stage 27.
It has a light source 26 for illuminating it with illumination light. Between the light source 26 and the wafer 28, the light source 2 is arranged in order from the light source 26 side.
Illumination lens group 32 that makes the illumination light from 6 parallel light, beam splitter 34 that splits the light reflected and returned by alignment mark 30, objective lens group 36, and illumination light that has passed through objective lens group 36. And a prism 38 that irradiates the wafer 28 perpendicularly to the wafer 28. Further, on the side where the beam splitter 34 splits the reflected light from the wafer 28, a CCD camera 44 for converting the received light into an electric signal is arranged via the reflecting mirror 40, the index mark 41 and the eyepiece lens group 42. ing. The CCD camera 44 is connected to a signal processing unit 46 that performs signal processing on the signal obtained by the CCD camera 44 and detects the position of the alignment mark 30. A monitor 48 that displays the waveform of the signal obtained by the CCD camera 44 is connected to the signal processing unit 46.

Further, in FIG. 3, the reduction projection lens 50 of the exposure apparatus for reducing and projecting the pattern of the reticle at the time of exposure in the lithography process is arranged in the vicinity of the area for forming the device pattern of the wafer 28. .

The illumination light emitted from the light source 26 is guided to the irradiation lens group 32. The irradiation lens group 32 includes one or a plurality of lenses, and makes the illumination light emitted from the light source 26 parallel light.

The illumination light passing through the irradiation lens group 32 passes through the beam splitter 34. Then, the illumination light transmitted through the beam splitter 34 irradiates the wafer 28 mounted on the stage 27 from the vertical direction via the objective lens group 36 including one or a plurality of lenses and the prism 38.

The light reflected by the alignment mark 30 of the wafer 28 passes through the objective lens group 36 via the prism 38, is reflected by the beam splitter 34, and is guided to the reflecting mirror 40. The light reflected by the beam splitter 34 sequentially passes through the index mark 41 and the eyepiece lens group 42 via the reflecting mirror 40. Then, an image is formed on the CCD element surface of the CCD camera 44.

The CCD camera 44 converts the received light into an electric signal and outputs it to the signal processing section 46. Signal processing unit 4
6 performs signal processing on the signal transmitted from the CCD camera 44, and detects the position of each alignment mark 30.

On the basis of the position information thus detected, the stage 27 of the exposure apparatus is driven and the wafer 28 is aligned.

In the alignment method according to the present embodiment,
Wafer alignment in the above-mentioned exposure apparatus,
The main feature is that it is performed based on the wafer magnification trend data stored in the database. Hereinafter, an example of the wafer magnification measuring method in the alignment method according to the present embodiment will be described with reference to FIG.

Alignment marks are formed in the arrangement shown in FIG. 4A on the wafer 52 which is aligned by the alignment method according to the present embodiment. Figure 4
In (a), the alignment mark is indicated by ● or ○. As shown in the drawing, the alignment marks B ₁ , B ₂ , B ₃ , B ₄ , B ₅ , B ₆ , B 7, B formed by the step portions are formed on the oxide film near the outer periphery of the wafer 52 from which the aluminum film is removed.
₈ ... is formed. In FIG. 4A, these are indicated by ●.

In each shot on the wafer 52, on the oxide film on which the aluminum film is formed and the resist is further formed, the alignment mark A ₁ composed of the step portion,
A ₁₁ , A ₂ , A ₃ , A ₄ , A ₄₄ , A ₅ , A ₆ , A ₇ , A ₈ , ...
... is formed. In the figure (a), these are indicated by a circle.

As described above, on the wafer 52, alignment marks are formed symmetrically in the vicinity of the outer periphery of the wafer 52 from which the aluminum film is removed and in each shot.

In measuring the wafer magnification, first, for example, the distance S _S between B _{1 in the} vicinity of the outer periphery on the left side and A ₁ in the left shot is measured. That is, the coordinates of A ₁ in the left shot with B ₁ as a reference are measured. S measured at this time
_S is measured as an error in the true distance S _s between B ₁ and A ₁ due to the asymmetric coating of A ₁ on the aluminum film. Therefore, considering the linear error and the non-linear error, S _S is expressed by the following equation using S _s .

S _S = αS _s −Δe Next, similarly, the distance S _L between B ₁ and A ₄ in the right shot is measured. That is, the coordinates of A ₄ in the left-side shot with respect to B ₁ are measured. S _L measured at this time
Is measured as an error in the true distance S _l between B ₁ and A ₄ due to the asymmetric coating of A ₄ on the aluminum film. Therefore, S _L is represented by the following equation using S _{1 in} consideration of linear error and non-linear error.

S _L = αS _l + Δe Here, the sign of the non-linear error Δe is expressed opposite to S _S because A ₁ and A ₄ are provided in the symmetrically positioned shots. .

Thus, the following equation is obtained by measuring the distance with B ₁ as a reference.

X ₁ = S _L −S _S = (αS _l + Δe) − (αS _s −Δe) = α (S ₁ −S _s ) + 2Δe Then, this time, within B ₄ near the outer periphery on the right side and in the right shot. Measure the distance S _S between A ₄₄ and. That is, the coordinates of A ₄₄ in the right shot with B ₄ as a reference are measured. S _S to be measured at this time, in the same manner as described above, considering the linear error and non-linear error, using S _s is expressed by the following equation.

S _S = −αS _s + Δe Then, similarly, the distance S _L between B ₄ and A ₁₁ in the right shot is measured. That is, the coordinates of A ₁₁ in the right shot with B ₄ as a reference are measured. Similarly, S _L measured at this time is represented by the following equation using S _{l in} consideration of linear error and non-linear error.

S _L = −αS _l −Δe Thus, the following equation is obtained by measuring the distance with B ₄ as a reference.

X ₂ = S _L −S _S = (− αS ₁ −Δe) − (− αS _s + Δe) = − α (S ₁ −S _s ) −2Δe The following equation is obtained.

(X ₁ −x ₂ ) / 2 = α (S ₁ −S _s ) + 2Δe In this way, the distance to the alignment mark in the shot is measured with reference to each wafer provided near the outer periphery of the wafer. Then, the same calculation is performed for each.

(X ₁ −x ₂ ) / 2 obtained with each of the alignment marks near the outer periphery as a reference is defined as x _i (i =
, 2, ..., N, ...), Δe is expressed by the following equation.

Δe = 1 / 2Σ (x _i −α (S _li −S _si )) ² Therefore, α that satisfies δ / δα (Δe) = 0 is obtained by the least square method.

Δ / δα {1 / 2Σ (x _i −α (S _li −S
_{From si} )) ² } = 0, α = Σx _i / Σ (S _li −S _si ). Therefore, α can be obtained from the obtained x _i and the measurement position (S _li , S _si ) (i = 1, 2, ..., N, ...) And the nonlinear component Δe can be derived. Here, the measured alignment mark has an error in the opposite direction on the left and right of the wafer, and interacts with α. In the measuring method described above, since the respective measurement results based on the left and right alignment marks are averaged, α can be accurately specified by the sampling effect.

The total sum and variations of Δe obtained as described above are stored in the database as trend data. FIG. 5 is a graph showing an example of trend data obtained as described above.

At the time of wafer alignment, the linear component of the positional deviation can be calculated based on the accumulated trend data by using the value obtained by subtracting each Δe from the measured value of the alignment mark position of each shot. By performing the wafer alignment based on the calculation result, it is possible to avoid erroneous detection of the position of the alignment mark formed on the underlying oxide film due to the coating with the aluminum film.

FIG. 6 is a schematic view showing an example of the apparatus configuration of an exposure apparatus when the above-mentioned alignment method is performed.

The exposure device 54 is connected to a storage device 56 which stores the wafer magnification trend data obtained as described above using the mark position detection system of the exposure device. The storage device 56 determines whether or not the wafer magnification needs to be corrected based on the trend data stored in the storage device 56, corrects the wafer magnification as necessary, and determines the wafer alignment in the exposure device 54. The device 58 is connected. The storage device 56 and the determiner 58 may be incorporated in the exposure device 54.

As described above, according to the present embodiment, the alignment mark is formed in the vicinity of the outer peripheral portion of the wafer and the aluminum film is removed so that the aluminum film or the like formed on the alignment mark can be easily removed. The wafer magnification is measured in an ideal state, the necessity of correcting the wafer magnification is judged based on the measured wafer magnification trend data, and the wafer magnification is corrected as necessary, so that the process is not complicated. Therefore, it is possible to avoid erroneous detection of the position of the alignment mark due to the asymmetric coating with the aluminum film. [Modified Embodiments] Various modifications are possible without being limited to the above-described embodiments of the present invention.

For example, in the above embodiment, the case where the aluminum film is formed on the alignment mark composed of the step portion formed in the insulating film has been described, but the case where the metal film other than the aluminum film is formed is also described. The invention can be applied.

Further, in the above embodiment, the alignment mark for removing the aluminum film is formed on the insulating film near the outer periphery of the wafer, but the position for forming the alignment mark for removing the aluminum film is an unnecessary position where the semiconductor device is not formed. If so, it is not limited to the vicinity of the outer periphery of the wafer.

[0064]

As described above, according to the present invention, the position of the first alignment mark, which is formed of the step portion formed on the insulating film on the wafer and on which the metal film is formed, is detected, The position of the reference alignment mark formed by the step portion formed on the upper insulating film and not having the metal film formed thereon is detected, and the first alignment is detected from the detection results of the positions of the reference alignment mark and the first alignment mark. Since the detection error of the position of the mark is obtained and the position information of the first alignment mark used for the wafer alignment is corrected in consideration of the detection error, the alignment mark of the alignment mark is increased without increasing the number of steps in the manufacturing process of the semiconductor device. It is possible to avoid erroneous position detection and realize highly accurate alignment in the exposure apparatus.

[Brief description of drawings]

FIG. 1 is a graph showing variations in wafer magnification and wafer rotation measured from an alignment mark covered with an aluminum film.

FIG. 2 is a schematic diagram showing the principle of the present invention.

FIG. 3 is a schematic diagram showing a configuration of an exposure apparatus.

FIG. 4 is a diagram illustrating a method for measuring a wafer magnification in an alignment method according to an embodiment of the present invention.

FIG. 5 is a graph showing an example of a wafer magnification trend chart measured by an alignment method according to an embodiment of the present invention.

FIG. 6 is a schematic view showing an apparatus configuration of an exposure apparatus when performing an alignment method according to an embodiment of the present invention.

FIG. 7 is a schematic view showing an aluminum film asymmetrically coated on an alignment mark formed of a step portion formed on an oxide film.

[Explanation of symbols]

10 ... Alignment mark 12 ... Wafer 14 ... Alignment mark 16 ... Wafer 26 ... Light source 27 ... Stage 28 ... Wafer 30 ... Alignment mark 32 ... Irradiation lens group 34 ... Beam splitter 36 ... Objective lens group 38 ... Prism 40 ... Reflector 41 ... Index mark 42 ... Eyepiece group 44 ... CCD camera 46 ... Signal processing unit 48 ... Monitor 50 ... Reduction projection lens 52 ... Wafer 54 ... Exposure device 56 ... Storage device 58 ... Judgment device 100 ... Wafer 102 ... Insulating film 104 ... Step 106 ... Metal film

   ─────────────────────────────────────────────────── ─── Continued front page    F term (reference) 2F065 AA03 AA21 BB02 BB27 CC19                       FF04 JJ03 JJ26 QQ18                 2F069 AA03 AA31 BB15 CC06 GG07                       NN17                 5F046 EA03 EA09 EB06 EC05 FC04

Claims (5)

[Claims] 1. A position of a first alignment mark, which comprises a step portion formed on an insulating film on a wafer and on which a metal film is formed, is detected and formed on the insulating film on the wafer, A step of detecting the position of the reference alignment mark formed of a step portion on which the metal film is not formed; and a step of detecting the position of the reference alignment mark and the position of the first alignment mark from the detection result of the first alignment mark. An alignment method comprising: a step of obtaining a position detection error; and a step of correcting the position information of the first alignment mark used for alignment of the wafer in consideration of the detection error. 2. The position of the first and second alignment marks having a step portion formed on an insulating film on a wafer and having a metal film formed thereon is detected, and the position of the insulating film on the wafer is detected. A step of detecting a position of a reference alignment mark formed of a step portion on which the metal film is not formed, and the step of detecting the position of the reference alignment mark and the positions of the first and second alignment marks First and second
And a step of correcting positional information of the first and second alignment marks used for alignment of the wafer in consideration of the detection error. Alignment method. 3. The alignment method according to claim 1, further comprising a step of removing a metal film formed on the reference alignment mark before detecting the position of the reference alignment mark. And alignment method. 4. A step formed on an insulating film on a wafer, a position of a first alignment mark having a metal film formed thereon is detected, and the step is formed on the insulating film on the wafer. Position detecting means for detecting the position of the reference alignment mark formed of a step portion on which the metal film is not formed, and the first alignment based on the detection results of the positions of the reference alignment mark and the first alignment mark. A detection error measuring unit for obtaining a detection error of the position of the mark; and a position information correction unit for correcting the position information of the first alignment mark used for alignment of the wafer in consideration of the detection error. And alignment device. 5. The position of the first and second alignment marks having a step portion formed on an insulating film on a wafer and having a metal film formed thereon is detected to detect the position of the insulating film on the wafer. Position detection means for detecting the position of a reference alignment mark formed of a stepped portion on which the metal film is not formed, and a detection result of the positions of the reference alignment mark and the first and second alignment marks. From the first and second
Error detection means for measuring the detection error of the position of the alignment mark, and position information correction means for correcting the position information of the first and second alignment marks used for the alignment of the wafer in consideration of the detection error. An alignment apparatus comprising: