Halbleiterwafer 1, bei dem Halbleiterelemente 3 mittels einer Maske auf den Wafer 1 belichtet werden, wobei die Justierung der Maske über einen Bezugspunkt auf der Maske und einen Referenzpunkt auf dem Wafer 1 erfolgt, deren räumliche Lage zueinander mittels eines Verfahrens zur Optimierung von Größen, die die Herstellungskosten eines Halbleiterelements 3 bestimmen, ermittelt wird.

A discrimination method usable in an exposure apparatus for transferring a pattern of an original (502) onto a substrate (103) is disclosed wherein information related to a relative positional deviation between the original and the substrate is produced, and wherein error shot discrimination is performed on the basis of the information and a condition which is changed in accordance with a layer or shot.

A projection exposure apparatus includes an illumination optical system for illuminating a first object (1) with exposure light (IL), a projection optical system (3) for projecting a pattern of the first object illuminated by the illumination optical system onto a second object (4), an alignment optical system (8-11) for illuminating the second object with alignment light (AL) of a wavelength different from the exposure light and for detecting alignment light from the second object through the projection optical system, and a moving mechanism for moving the first object to a position (A), different from a position (B) of the first object for the projection exposure, so that the alignment optical system detects the alignment light from the second object through the projection optical system.

When carrying out the alignment process, light from a first and a second irradiation device (1,6) illuminates alignment marks (MAM,WAM) of a mask (M) and a workpiece (W) respectively, the projected images of the alignment marks are recorded, subjected to image processing and relative positions of the corresponding alignment marks are determined and stored. The relative positions of the mask and workpiece alignment marks are computed and the workpiece and/or the mask are/is moved such that the corresponding marks come to rest one on top of the other.

Alternatively, actinic light is used to illuminate the alignment marks of a mask and projected onto a reflecting component which is located on the workpiece carrier. After stopping emission of the actinic light, the workpiece is placed on the workpiece carrier and non-actinic light illuminates the alignment marks of the workpiece. The relative positions of the corresponding alignment marks are determined and stored and the workpiece and/or the mask are/is moved such that the corresponding alignment marks come to rest one on top of the other.

Alignment correction coefficients are calculated by a method of least squares with reference to the coordinates of each of the alignment marks and the coordinates of the rational grid points. Corrected coordinates of the alignment marks are derived from the coordinates of the rational grid points on the basis of the alignment mark correction coefficients, each of their statistical functions are calculated, and the differences of these coordinates are calculated. The residuals at each of the alignment marks are calculated with reference to the difference between the alignment displacement tolerance values in the predetermined X and Y directions and their coordinates. The corrected coordinates of the alignment marks associated with the first pattern for optically exposing the second pattern from the alignment mark correction coefficients and their random number elements when the residual sum of squares becomes a minimum value. In this way, optimal correction of the alignment can be carried out even in the case that the random elements in the alignment mark displacement components from the rational grid point do not follow a normal distribution.

A register mark and a system for bringing a pair of register marks into alignment is disclosed. A first two-dimensional register mark (Fig. 1) comprises a plurality of dots of a first frequency, and a second two-dimensional register mark (Fig. 2) comprises a plurality of dots of a second, generally higher frequency. When the first and second register marks are overlaid, an interference pattern resulting from the difference in frequency is observed. The first and second register marks are in alignment when the interference pattern produces a maximum bright spot in the center of the register mark. A small relative movement of the first and second register marks results in a larger relative movement of center of the bright spot. The position of the bright spot indicates which direction and how much to move the second register mark in order to achieve image registration. It is preferred that the frequency of dots in each register mark is warped with distance from the center of the mark. The frequency of the dots of the second register mark is also warped with distance from the center, but overall has a generally higher frequency of dots as compared to the first register mark. In the resulting interference pattern the bright spot occurs unambiguously when the two register marks are in alignment. At misalignments of more than one dot width, the bright spot breaks up so that the overlaid register marks may serve as quick visual check of proper alignment. The register marks may be in a single dimension along a line.

An alignment apparatus for use in an exposure system for exposing fine patterns on a wafer, the alignment apparatus comprising a light source optical system for emitting coherent alignment light, a positional deviation detecting optical system for receiving the alignment light reflected from the wafer, and a light-receiving optical system for detecting a positional deviation of the wafer on the basis of the alignment light received by the positional deviation detecting optical system. These three optical systems are arranged to be coupled through flexible optical fibers to each other. This coupling arrangement using the flexible optical fiber can reduce the size of the positional deviation detecting optical system whereby the positional deviation detecting optical system can be disposed directly under a projection lens of the exposure system, thereby accurately effecting the alignment of the wafer with respect to the projection lens.

A method is described for repetitively imaging a mask pattern, on separate fields of a substrate (W), for example, for IC manufacture, which substrate fields are positioned without any field-by-field alignment so that the speed of throughput of substrates can be increased. An accurate interferometer system (50, 100, 150) having five measuring axes (MAX₁, MAX₂, MAX₃, MAX₄, MAX₅) is also described, which system is intended in the first instance for use in an apparatus for performing the method, but which can also be used in a more general way in those cases where an object must be measured in five degrees of freedom.

An alignment method for use in an exposure apparatus for printing a pattern of an original (2) onto different surface areas of a substrate (3), the alignment method comprising the steps of: providing alignment marks around the pattern of the original and placing the original on an original supporting stage (4); providing a reference mark (14) on an X-Y stage (24) for supporting the substrate and being movable in X and Y directions, and moving the X-Y stage so as to place the reference mark at those positions, in sequence, which correspond to the alignment marks of the original, respectively, and which are preset in respect to a stage coordinate system; detecting, in sequence, positional errors of the alignment marks of the original with respect to the corresponding set positions, respectively, by using the reference mark and through the movement of the X-Y stage, wherein the positional errors are detected by use of positional error detectors (12) which are provided to be associated with the alignment marks of the original, respectively; calculating a rotational error of the original with respect to the stage coordinate system, in ϑ direction, by using the positional errors; and rotationally moving the original supporting table in the ϑ direction so as to correct the rotational error.

This invention relates to a lens system having chromatic aberration on the axis of a double-focus detector which utilizes chromatic aberration. One focal plane of the lens system consists of a light image forming plane of a short wavelength corresponding to a first object which is, for example, a mask and the other focal plane consists of the same image forming plane of a wavelength having a predetermined band corresponding to a second object which is, for example, a wafer. The first object is illuminated with the light of a wavelength shorter than 500 nm and the second object is illuminated in a predetermined band of wave-lengths greater than 500 nm.

An exposure apparatus for transferring a pattern of an original (1, 2) onto a workpiece (3), includes a blocking member (7a-7d) for defining a rectangular exposure region with respect to at least one of the original (1, 2) and the workpiece (3), wherein the exposure for the pattern transfer can be effected with the exposure region defined by the blocking member (7a-7d). Plural detection systems detect a positional deviation between the original (1, 2) and the workpiece (3), each of which is disposed so as to be associated with at least one of four sides of the rectangular exposure region. Plural first movable stages (8a-8d) each is provided so as to be associated with at least one of the four sides, and each is adapted to carry thereon one of the detection systems disposed to be associated with a corresponding side. Each first movable stage (8a-8d) comprises a single-axis stage movable in a direction parallel to a corresponding side. Plural second stages (9a-9d) each carrys thereon corresponding one of the first movable stages (8a-8d), and each comprises a single-axis stage movable in a direction perpendicular to a corresponding side and in a direction parallel to the rectangular exposure region. Each second movable stage (9a-9d) is operable to displace the light blocking member (7a-7d) to change the rectangular exposure region.

A double-focus detector utilizing chromatic aberration vertically illuminates a mask (9) and a wafer (10) which are disposed at a minute interval in the direction of illumination rays, in the direction normal to the mask and the wafer through a band-pass filter (6) or band-pass filters which transmit either both of a single wavelength ray and a wavelength-band ray, and an objective (8) to produce axial chromatic aberration corresponding to the minute interval with respect to the single wavelength ray and the wavelength-band ray, and observes images of the mask and the wafer which are formed by the single wavelength and the wavlength-band ray reflected by the mask and the wafer, thereby detecting a relative position between the mask and the wafer in a direction perpendicular to the optical axis of the illumination rays.

A mask structure according to an aspect of the invention includes a base having a first zone for an alignment pattern and a second zone for a circuit pattern; an alignment pattern forming material with which the alignment pattern is formed in the first zone; and a circuit pattern forming material with which the circuit pattern is formed in the second zone; wherein the alignment pattern forming material and the circuit pattern forming material are different from each other.

A double-focus detector utilizing chromatic aberration vertically illuminates a mask (9) and a wafer (10) which are disposed at a minute interval in the direction of illumination rays, in the direction normal to the mask and the wafer through a band-pass filter (6) or band-pass filters which transmit either both of a single wavelength ray and a wavelength-band ray, and an objective (8) to produce axial chromatic aberration corresponding to the minute interval with respect to the single wavelength ray and the wavelength-band ray, and observes images of the mask and the wafer which are formed by the single wavelength and the wavlength-band ray reflected by the mask and the wafer, thereby detecting a relative position between the mask and the wafer in a direction perpendicular to the optical axis of the illumination rays.

Described is a method for facilitating the alignment of a photomask with individual fields on the surfaces of a number of wafers onto which fields the pattern of the photomask is to be imaged for projection printing, each of said number of wafers carrying an identical array of fields produced in at least one previous printing process. For this purpose it is possible to chose one out of the identical batch of wafers, measure the rotational adjustment necessary to being each field into the correct direction in the horizontal plane and to use the measured value for an automatic rotational adjustment of all other wafers of the batch.