CN1374561A - Exposure method and exposure apparatus - Google Patents

Abstract

An exposure method suitable for bonding patterns (A, B) on a substrate (P) to expose desired patterns, the method is to perform a preliminary exposure process to expose patterns (A, B) on the substrate different A first mark (MK) and bonding patterns (A, B). Then, a setting process is carried out, the setting process is based on the relative positional relationship between the exposed first mark (MK) and the pattern (A, B) on the substrate to set the exposure bonding pattern (A, B) correction amount.

Description

Exposure method and exposure device

technical field

The present invention relates to an exposure method and an exposure device for exposing a mask pattern on a substrate in the process of a semiconductor element or a liquid crystal element, and in particular relates to an exposure method and an exposure device in which identical parts of a plurality of partition patterns are bonded to each other. A large-area pattern is exposed on the substrate, which is an exposure method and an exposure device suitable for combining screens.

Background technique

A known exposure apparatus uses a so-called screen combining method in order to cope with an increase in the size of a photosensitive substrate to be exposed. This method divides the exposure area of the photosensitive substrate into a plurality of unit areas (the area irradiated by each exposure (Shot)), and then repeats the exposure area of each unit for multiple exposures to combine the pattern into a Ultimately desired large area pattern. In the case of combining screens, due to the pattern error of the grating (Reticle) for pattern projection, the distortion (Distortion) of the projection optical system, and the position determination error of the stage (Stage) that determines the position of the photosensitive substrate, etc., each There may be gaps in the pattern joined at the boundary position of the unit exposure area. Here, in order to prevent the occurrence of pattern gaps, the boundaries of the unit exposure areas are slightly overlapped, that is, the screens are merged by overlapping the unit exposure areas.

When performing the above-mentioned picture merging, the most important thing to pay attention to is that the offset between the patterns in the picture merging part must be minimized to ensure the accuracy of the picture joining. The inter-pattern misalignment in the screen merged part is caused by manufacturing error of the grating, lens aberration of the projection optical system, position determination accuracy and sliding accuracy of the stage that determines the position of the photosensitive substrate and other materials. That is, in the above-mentioned frame merging, the relative positional deviation between two adjacent patterns will cause a drop in the joint portion of the patterns, which will damage the characteristics of the manufactured components. Moreover, in the manufacture of semiconductor elements and liquid crystal display elements, since the single-layer patterns of screen integration are overlapped into multiple layers (usually 5 to 8 layers in the manufacture of liquid crystal panels), the overlap error of the unit exposure area in each layer It will cause discontinuous changes in the joint part of the pattern. In this case, especially for active matrix (Active Matrix) liquid crystal elements, the discontinuous change of contrast (Contrast) in the pattern joint part will reduce the quality of the element.

Generally, in terms of the splicing accuracy of the exposed layers after the second layer, since it is important to know the overlap accuracy difference between the exposed shot area and the exposed shot area, it is possible to first measure the alignment marks formed on the first layer ( Alignment Mark), and then calculate the deviation (Offset) of the grating and photosensitive substrate of the second and subsequent layers from the measurement results for correction. That is, when performing bonding exposure on the second layer (for example, a source/drain layer), since the alignment mark formed in the first layer (for example, a gate layer) that is the previous layer has a standard, it is Offaxis (Offaxis) and other detectors (Sensor) measure this alignment mark, which can increase the overlapping of the source and drain layer patterns and their respective corresponding gate layer patterns to a high degree of accuracy. As a result, the second layer Patterns are joined with high accuracy.

However, the bonding accuracy of the patterns in the second layer and later is affected by the bonding accuracy of the patterns in the first layer. In the general exposure process, when the screen bonding is performed in the first layer, since the substrate has no reference, the exposure step will be affected by the mechanical accuracy or the manufacturing accuracy of the pattern. Therefore, it is necessary to increase the bonding accuracy to a high accuracy. very difficult. In view of this, it is known to ensure the bonding accuracy in the first layer by, for example, the following two methods.

(1) In process

Regarding the manufacturing error of the grating or the lens aberration of the projection optical system, as shown in FIG. As shown, specifically, the plurality of special patterns near the pattern A in the grating are a plurality of two-dimensional marks arranged along the joint edge, so the above-mentioned The position of the mark, and the deviation error between the design position of each mark and the position of the measurement result is statistically processed, and the correction parameter (Shift) that can make the pattern on the grating exposed on the carrier table to a more ideal grid shape can be obtained. , rotation, magnification, etc.). In addition, as a method of measuring the position of the mark from the vicinity of the stage, a method such as measurement by a so-called image slit sensor (ISS) can be used. In this method, a fiducial mark member with a slit mark (Slit Mark) is installed on the stage. , and then irradiate detection light with the same wavelength as the exposure wavelength from below the reference mark member, and monitor the detection light incident on the light quantity detector through the aperture mark, the projection optical system, and the mark on the grating while scanning the moving stage to correspond to The stage coordinate system calculates the position of the marker.

Moreover, as for patterns B to D, it is also possible to arrange marks along the joint edge like pattern A, and then use the position of the measurement marks to obtain correction parameters that can expose patterns B to D to a more ideal lattice shape. In addition, in terms of the positioning accuracy or sliding accuracy of the stage that determines the position of the photosensitive substrate, errors can be minimized during adjustment to maintain the screen joint accuracy.

(2) Test exposure

Before implementing the process exposure, first implement the bonding exposure of combining the patterns A to D to form the desired pattern, and measure the bonding part with a measuring device to obtain the position offset Z as shown in Figure 12, so that the position offset The shift Z is used as a deviation from the design value, and as a method of inputting the deviation in the raster portion of the exposure control data during the process exposure.

However, the above known exposure method and exposure apparatus have the following problems.

In the technology of (1) above, because it is impossible to measure the transferred patterns A to D, the pattern after exposure cannot be measured, and the optical principle measurement such as ISS measurement is used, even if the exposure is performed according to the correction parameters, the actual The pattern after exposure also produces deviations from the designed values. Among them, there are several reasons to consider, such as the basic error in the measurement principle or the limit of reproducibility, the influence caused by the deformation of the photosensitive substrate in the process conditions, the manufacturing error of the actual pattern and the mark for measurement, etc.

Moreover, in the technique of (2) above, in the case of combining images in the above-mentioned first layer, it is difficult to enhance the bonding accuracy because there is no reference in the base, so even if a sensor is additionally provided on the first layer to The method of detecting a position of the exposure pattern, and then exposing other patterns according to the measured pattern position, since the shape of the pattern usually extends in a single direction such as the X direction or the Y direction in each layer, it is necessary to start from X, It is very difficult to measure a position of the above pattern in both Y directions. Therefore, for directions that cannot be measured (X direction or Y direction), it is impossible to sufficiently improve the joining accuracy.

For the reasons described above, the techniques (1) and (2) above are not sufficient as methods for improving the joining accuracy.

Contents of the invention

An object of the present invention is to provide an exposure method and an exposure apparatus, which can accurately improve bonding accuracy when bonding patterns on a substrate for screen combination.

In order to achieve the above object, the embodiments of the present invention employ structures corresponding to FIGS. 1 to 12 installed.

The exposure method of the present invention is suitable for bonding patterns (A-D) on a substrate (P) to expose desired patterns, and is characterized in that the method includes performing a pre-exposure process to expose patterns (A-D) on the substrate. ) different first marks (MK) and bonding patterns (A-D), and perform a setting process, which is based on the relationship between the first marks (MK) and the patterns (A-D) exposed on the substrate Set the correction amount when exposing joint patterns (A to D) according to the relative positional relationship between them.

The exposure device of the present invention is suitable for bonding patterns (A-D) on a substrate (P) to expose desired patterns, and is characterized in that the device includes a storage device (33) and a correction device (23). The storage device (33) can store the exposed bonding patterns (A-D) according to the relative positional relationship between the exposed first marks (MK) different from the patterns (A-D) and the patterns (A-D) exposed on the substrate The correction amount of time. The correction device (23) can bond patterns on the substrate (P) according to the correction amount stored in the storage device (33).

Therefore, in the exposure method and exposure apparatus of the present invention, for example, the two-dimensional marks extending in the X direction and the Y direction in the past are used as the first mark (MK), and the patterns (A-D) are exposed to the substrate (P) at the same time. Above, from the measurement of the exposed patterns (A to D) and the first mark (MK), the pattern (A to D) bonding deviation as a two-dimensional correction amount can be obtained. Therefore, at the time of actual exposure, the joining accuracy can be sufficiently improved by joining the patterns in a state corrected by this correction amount. Furthermore, since the present invention uses an actual transfer pattern when calculating the correction amount, the necessary correction amount can be directly calculated during actual exposure.

Description of drawings

FIG. 1 is a schematic configuration diagram of an exposure apparatus according to an embodiment of the present invention.

FIG. 2A is a plan view of a grating with marks used in the exposure method of the present invention.

Figure 2B is a detailed view of the labeling.

3 is a plan view of a plurality of patterns and marks exposed on a substrate.

FIG. 4A is a partially enlarged view of a gate pattern.

FIG. 4B is an enlarged view of the overlapping portion of the gate pattern and the mark.

FIG. 5 is a partially enlarged view of a gate pattern and a mark exposed across a dividing line.

FIG. 6 is a schematic diagram of the relative positional relationship between an ideal grid and a pattern.

FIG. 7 is an enlarged view of an overlapping portion of a mark in another shape and a gate pattern.

Fig. 8 is an explanatory diagram of another method of exposing marks.

FIG. 9 is a flowchart of an example of a liquid crystal display (semiconductor) element process.

FIG. 10 is a schematic diagram of bonding a plurality of patterns.

Fig. 11 is a plan view of patterns and two-dimensional marks formed on a grating.

FIG. 12 is a schematic diagram of deviations occurring in bonding patterns.

Explanation of reference signs:

9: Exposure device

10: light source

11: Illumination optical system

12: Projection optical system

13: grating carrier

14: substrate carrier

15: Momentary switch

16, 17: Mirror

18: Wavelength Selective Filter

19: Optical integrator

20: variable field aperture

21: Concentrator optical system

22: Momentary switch device drive unit

23: Control device

24: Beam Splitter

25: Grating carrier drive system

26a, 26b: Grating alignment system

27: Carrying platform driving device

28: Moving Mirror

29: Laser interferometer

30a, 30b: auto focus system

32: Standard marker component

33: storage device

34: Mirror

35: Concentrating lens

36: optical fiber

37: Light sensor

38: Scan line pattern

39: Gate electrode pattern

40: area

201, 202, 203, 204, 205, 206: steps

A, B, C, D: pattern

DLX, DLY: dividing line

K: ideal lattice

LX, LY, PX, PY: spacing

MK: mark

OVLx, OVLy, OVRx, OVRy, OVUx, OVUy, OVDx, OVDy: Overlap deviation

P: Substrate

R, RT: Raster

XM: X mark

YM: Y mark

Z: Offset

Detailed ways

Please refer to FIG. 1 to FIG. 9 , which are used to illustrate embodiments of the exposure method and exposure device of the present invention.

Here, the case where a plurality of grating patterns are merged on a screen using a vertical equal magnification on a glass substrate (photosensitive substrate) for manufacturing a liquid crystal display element (hereinafter referred to as a substrate) as an example is taken as an example. illustrate. Components in FIGS. 1 to 9 that are the same as those in known FIGS. 10 to 12 are denoted by the same reference numerals, and description thereof will be omitted.

FIG. 1 is a schematic configuration diagram of an exposure apparatus 9 for manufacturing a liquid crystal display element. This exposure device 9 is a device for projecting and exposing a substrate (substrate) P coated with a photosensitive agent (photoresist) to project a liquid crystal display element pattern formed in a grating (mask) R. This device is generally composed of a light source 10, an illumination An optical system 11 , a projection optical system 12 , a grating carrier (mask carrier) 13 and a substrate carrier (substrate carrier) 14 are formed. Among them, the Z axis is set to be parallel to the optical axis of the projection optical system 12, the X axis is parallel to the paper surface of Fig. 1 in the vertical plane of the optical axis, and the X axis is parallel to the paper surface of Fig. The vertical plane is the Y axis.

The light source 10 is a device that generates a light beam B as exposure light, and this device is composed of an ultra-high pressure mercury lamp or the like. The light beam B emitted from the light source 10 enters the illumination optical system 11 .

The illumination optical system 11 is composed of a momentary switch (Shutter) 15 that can switch the light path of the light beam B, reflectors 16, 17, a wavelength selection filter 18, and an optical integrator 19 (Optical Integrator) that makes the light beam B uniform ( Fly-Eye Lens (Fly-Eye Lens)), variable field aperture 20, condenser (Condenser) optical system 21 and so on. The momentary switch device 15 is controlled by the control device 23 (correction device) passing through the momentary switch device drive unit 22 to drive the momentary switch device 15 to switch the light path of the light beam B. Then, the light beam B incident to the illumination optical system 11 will correspond to the opening action of the momentary switch device 15, select the wavelength (g-line or h-line, I-line) required for exposure in the wavelength selection filter 18, and use the optical product Calculator 19 makes the brightness uniform. After the light beam B with uniform brightness passes through the beam splitter (BeamSplitter) 24, it is concentrated by the condenser optical system 21, and passes through the opening of the variable field diaphragm 20 to overlap and illuminate the illumination area on the specified grating R . The position and size of the opening of the variable field diaphragm 20 can be controlled by a control device 23 through a blind driving unit (not shown).

The grating stage 13 is a device for holding the grating R, which can be driven by the grating stage driving system 25 and can move two-dimensionally on the XY coordinate system. Furthermore, above the grating stage 13, photoelectric sensors as grating alignment systems 26a and 26b are arranged. The grating alignment systems 26a and 26b are devices capable of irradiating alignment light with the same wavelength as the light beam B emitted by the light source 10, and the reflected light can be processed by the CCD camera for image processing. Then, the exposure device can detect the alignment mark on the grating R formed by chrome or the like in a linear shape, and then use the grating stage driving system 25 to drive the grating stage 13 according to the detection result, so that the grating R is aligned (compared to the position ) The position set by the XY coordinate system.

The projection optical system 12 forms an image of a pattern existing in the illumination area of the grating R on the substrate P. As shown in FIG. Then, after the photosensitizer coated on the substrate P is exposed to light, the image of the pattern is transferred to the substrate. The imaging characteristics (magnification, etc.) of this projection optical system 12 can be adjusted by the control of the control device 23 .

The substrate stage 14 is a device for holding the substrate P, which can be driven by the stage driving system 27 and can move two-dimensionally on the XY coordinate system. Since the substrate carrier 14 is equipped with a moving mirror 28, the laser light emitted from the laser interferometer 29 can be reflected by the moving mirror 28 and returned to the laser interferometer 29, so the interference between the reflected light and the incident light of the laser light can be accurately measured. The position of the substrate stage 14 (the position of the substrate P), and the measurement result of the laser interferometer 29 will be output to the control device 23 . In addition, for the sake of simplicity in describing this embodiment, only the moving mirror 28 and the laser interferometer 29 for measuring the position of the substrate stage 14 in the X direction are shown in FIG. Moving mirror and laser interferometer.

Between the projection optical system 12 and the substrate stage 14, oblique incidence autofocus systems 30a, 30b are provided. The position set by the direction of the optical axis. That is, the substrate stage 14 is driven in the Z direction so that the exposed surface of the substrate P coincides with the focal plane of the projection optical system 12 .

Further, on the substrate stage 14 , a disk-shaped standard marker member 32 is provided at a position where the direction of the optical axis of the projection optical system 12 roughly coincides with the exposed surface of the opposing substrate P. As shown in FIG. Aperture markers (not shown) opened in a rectangular shape are provided in this standard marker member 32 . Then, a mirror 34 and a condenser lens 35 are installed below the standard marking member 32 in the substrate stage 14, and the light beam B (exposure light) as detection light transmitted by the optical fiber 36 will pass through the condenser lens 35 and the mirror from below. 34 while the standard marking member 32 is illuminated.

The image of the aperture mark on the illuminated standard mark member 32 is back-projected onto the grating R through the projection optical system 12 . Then, the light beam B transmitted through the grating R passes through the condenser optical system 21 and the mirror 18 and enters the beam splitter 24 . The light beam B entering the beam splitter 24 is reflected inside the beam splitter 24 and then enters a photoelectric sensor as a light quantity sensor 37 . The light sensor 37 outputs an electrical signal to the control device 23 corresponding to the intensity of the incident light beam B. Furthermore, the light quantity sensor 37 is arrange|positioned on the surface which couples with the grating R. As shown in FIG.

The control device 23 uses the signal output from the light sensor 37 and the signal output from the laser interferometer 29 to detect the position of the mark on the grating, and simultaneously uses the deviation between the detected position of the mark and the design position to perform calculation processing ( Least square method) to calculate the correction parameters such as the rotation correction amount of the grating R, the XY offset correction amount, and the XY magnification deviation amount. Then, the control device 23 judges the position of the grating R through the grating stage driving system 25 according to the calculated correction amount, and adjusts the imaging characteristics of the projection optical system.

Moreover, the control device 23 can collectively control the above-mentioned projection optical system 12, the momentary switch device driving part 22, the shielding driving part, the grating stage driving system 25, and the stage driving device 27; detect the position of the substrate stage 14; The substrate stage 14 two-dimensionally moves the substrate P to expose and form a set pattern obtained by bonding the grating R pattern on the substrate P. In addition, the control device 23 is provided with a storage device 33 capable of recording various exposure data (prescriptions) such as program (Sequence) parameters or correction parameters.

Next, the processing (preliminary exposure process) performed before the process processing will be described. The pre-exposure process can generally be divided into a pattern exposure process in which the gate pattern to be exposed is exposed on the substrate P, and a mark exposure process in which the mark MK is exposed on the substrate P on which the gate pattern has been exposed. In addition, as shown in FIG. 10 , in the first layer of the substrate P, the gate layer (gate pattern) formed by bonding the patterns A to D is a pattern formed by combining screens.

As shown in FIG. 12 , each pattern A to D has a scanning line pattern (unit pattern) 38 extending in the X direction, and a plurality of gate electrodes arranged at a fixed pitch LX (for example, a pitch of 100 μm) along the scanning line pattern 38 The pattern (unit pattern) and the same gate pattern arranged at a constant pitch LY in the Y direction are usually set so as to intersect the scanning line pattern with the dividing line joining the patterns A to D. Moreover, after the gate pattern is developed on the exposed substrate P, the photoresist with the above gate pattern will remain on the substrate, that is, a positive pattern is formed in the grating R.

On the other hand, FIG. 2A is a raster RT for test exposure.

As shown in FIG. 2A, a mark MK as a first mark is formed near the center of the raster RT. This mark MK is different from the above-mentioned gate pattern or other patterns (source/drain patterns) formed on the substrate P in the process. As shown in FIG. 2B , each mark MK is composed of an X mark XM and a Y mark YM. The X mark XM extends to the Y direction and is used to measure the position in the X direction. The Y mark YM extends in the X direction and is used for measuring the position in the Y direction.

The arrangement pitch PX of the X marks XM is set to 1/n (n is a natural number) of the arrangement pitch LX of the gate electrode patterns 39 , where PX=LX (ie, n=1). The arrangement pitch PY of the Y marks YM is set to be 1/m (m is a natural number) of the arrangement pitch LY of the scanning line pattern 38 , where PY=LY (that is, m=1). Moreover, after the mark MK is developed on the exposed substrate P, the photoresist having the shape of the mark MK can be removed from the substrate P, that is, a negative film pattern is formed in the grating R.

Therefore, in the initial pattern exposure process, the grating R having the gate pattern is used to expose the gate pattern in the first layer on the substrate P. Referring to FIG. At this time, as shown in FIG. 3 , while replacing the four gratings each having the patterns A to D, the exposed patterns A to D are joined sequentially by the dividing line DLY extending in the Y direction and the dividing line DLX extending in the X direction. , and a window pane pattern is formed on the substrate P through screen integration.

Furthermore, when a plurality of gratings having patterns A to D are used to form the pane pattern, the variable field diaphragm 20 can be adjusted by utilizing the shape characteristics of the gate pattern or the source/drain pattern being repeatedly arranged at a certain pitch. The illuminated area on the raster acts as each pattern that changes sequentially. In this case, the number of gratings can be reduced, and the grating exchange time can be reduced, thereby improving production efficiency.

Next, set the grating RT for test exposure on the grating stage 13, move the substrate stage 14 sequentially, and expose the four marks MK along the center of the substrate P and the dividing lines DLY and DLX multiple times (9 times in FIG. 3 ). ). When the mark MK is exposed using the grating RT for test exposure, it is preferable that the substrate P can be aligned, and it can be continuously implemented after the gate pattern exposure. Therefore, it is possible to prevent the test exposure result from including errors caused by substrate alignment and error factors other than bonding accuracy.

In the mark exposure process, the mark MK is exposed in two patterns across the dividing line DLY (and/or DLX) in each exposure. That is, the mark MK can be simultaneously exposed on two patterns adjacent to each other on the substrate by one irradiation. Also, each mark MK may correspond to the gate pattern formed in the pattern exposure process shown in FIG. 4A. As shown in FIG. 4B , the X mark XM overlaps the center position of the gate electrode pattern 39 in the X direction in the gate pattern, and the Y mark YM overlaps the center position in the Y direction of the scan line pattern 38 in the gate pattern.

In addition, in the grating R, since the mark MK is arranged near the center of the grating, that is, near the optical axis of the projection optical system 12, it is possible to reduce the grating components caused by the magnification or rotation (Rotation) when projecting through the projection optical system 12. error.

Then, after exposing the gate pattern and the mark MK on the substrate P through a pre-exposure process, the substrate P is subjected to a developing treatment. Here, since the gate pattern is a positive film pattern, the light beam B passing through the region other than the gate pattern on the grating R, as shown in FIG. , and the regions corresponding to the gate patterns (patterns 38, 39) become unexposed regions. In addition, since the mark MK is a negative film pattern, as shown in FIG. 4B, the mark MK is formed by exposure on the patterns 38, 39 which are unexposed regions. Therefore, even if the substrate P after the pre-exposure process is subjected to development treatment, the mark MK can be formed in the gate pattern (however, in FIG. remove). Furthermore, if the latent image measurement can be performed, it is not necessary to develop the substrate P, and the following measurement step after the pre-exposure process can be directly performed.

After the pre-exposure process, the relative positional relationship between the gate pattern transferred to the substrate P and the mark MK can be measured using, for example, an overlay measuring device. Specifically, as shown in FIG. 5 , for example, the relative positional relationship between the gate pattern and the mark MK joined in the X direction via the dividing line DLY can be measured on both the pattern A and B sides.

As shown in FIG. 5, in the case where both patterns A and B are loaded into one pane (Window) to detect an edge (Edge), the arrangement pitch LX of the gate electrode patterns 39 (as shown in FIG. 12 ), due to The pixel pitch is about 300 μm, and the RGB division is about 100 μm, so the detection magnification of the objective lens in the measuring device must be greatly reduced. However, a reduction in the detection magnification will also reduce the detection resolution, resulting in a reduction in detection accuracy. Therefore, in this embodiment, by measuring the relative positional relationship between each gate pattern of the patterns A and B and the mark MK, it is possible to fully detect the entire area of tens of μm, and in the state of increasing the detection magnification of the objective lens , which can perform edge detection within a pane.

Hereinafter, the relative position measurement will be described next.

For example, in FIG. 5 , the overlap deviation between the gate pattern and the mark MK in the irradiation area (pattern A) on the left side, the overlap deviation between the gate electrode pattern 39 and the X mark XM in the X direction is OVLx, The amount of overlap deviation between the scanning line pattern 38 and the Y mark YM in the Y direction is OVLy. Similarly, the overlap deviation between the gate pattern and the mark MK in the irradiation area (pattern B) on the right side, the overlap deviation between the gate electrode pattern 39 and the X mark XM in the X direction is OVRx, and the scanning line in the Y direction The amount of overlap deviation between the pattern 38 and the Y mark YM is OVRy.

Here, the mark MK is formed on a grating, because the distance between the position of the mark and the mark can be measured in advance, and the reliability of the mark MK is higher than that of the gate pattern on the substrate P. Therefore, the above-mentioned amount of overlay deviation can be based on the position of the mark MK, and the overlay deviation and joint deviation (joint error) of the gate pattern corresponding to the mark MK can be found.

Therefore, the bonding accuracy between the pattern A and the pattern B can be expressed by the following formula using the above-mentioned overlay deviation amount.

X direction: |OVLx-OVRx|

Y direction: |OVLy-OVRy| (1)

Therefore, it is known that it is difficult to find out the joining accuracy of both the X direction and the Y direction, but it can be easily found here by using another easy method.

In addition, in addition to the above-mentioned joint accuracy between patterns A and B, for example, the joint accuracy between patterns A and C in the Y direction is obtained (not shown), and both patterns A and C are obtained in the same procedure as above. The relative positional relationship between the gate pattern and the mark MK joined in the Y direction across the dividing line DLX was measured. Therefore, in the upper irradiated area (pattern A), the amount of overlap deviation OVUx between the gate pattern and the mark MK in the X direction and the amount of overlap deviation in the Y direction as OVUy are obtained, while in the lower irradiated area (pattern C) Calculate the overlap deviation amount of the gate pattern and the mark MK in the X direction as OVDx and the overlap deviation amount in the Y direction as OVDy. The joining accuracy between the pattern A and the pattern C was calculated|required by the following formula based on formula (1).

X direction: |OVUx-OVDx|

Y direction: |OVUy-OVDy|

Then, the amount of overlap deviation between the gate pattern and the mark MK is measured for each of the other regions by the same procedure as above. Then, as shown in FIG. 6, for example, the correction amount when exposing the pattern A, that is, the correction parameters (offset X, Y, rotation, magnification, etc.) at the time of exposure using the grating R having the pattern A are in complex Among the marks, the least square method is used to obtain the overlap deviation and design value of the plurality of marks formed along the joint edge of the pattern A and the gate pattern through statistical calculation. Similarly, correction parameters for patterns B to D during exposure are obtained by using the amount of overlap deviation between a plurality of marks formed along the joint sides of each pattern and the gate pattern.

The correction parameters described above can correct errors including pattern errors of the grating R, distortion of the projection optical system, and errors calculated using the grid pattern actually exposed on the substrate P. FIG. Therefore, each correction parameter for each pattern obtained in the measurement step can be recorded in the storage device 33 .

After the preliminary exposure process and the measurement process are completed, the exposure process is performed. For example, the grating with pattern A used in the pre-exposure process is transported to the grating carrier 13 by an unshown optical transfer system, and then the control device 23 first measures the illumination formed on the grating R with the grating alignment system 26a, 26b The grating mark (not shown) outside the area, and the alignment of the grating R itself is performed through the grating carrying table driving system.

Next, the control device 23 uses the autofocus systems 30 a and 30 b to position the fiducial mark member 32 and the substrate P at positions corresponding to the grating R in the optical axis direction of the optical projection system 12 .

Next, in the setting process, the control device 23 reads the pre-obtained rotation correction amount, XY displacement correction amount, and XY magnification deviation amount of the grating R from the storage device 33, and then determines through the grating stage 13 according to the correction parameters. position of the grating R, and adjust the imaging characteristics of the projection optical system 12 at the same time. Therefore, as shown in FIG. 6, the pattern A of the grating R can be corrected for deviations generated in the ideal lattice K. FIG.

After aligning corresponding to the grating R, the exposure irradiation area can be determined on the area on the substrate P where the pattern A is exposed, and then the substrate carrier 14 is driven through the carrier driving device 27, and the grating R is illuminated by the light beam B, so that On the first layer of the substrate P, the pattern A as a part of the gate pattern is exposed. Then, carry out the same procedure as pattern A to correspond to the gratings respectively having patterns B～D, use the correction parameters read from the storage device 33 to align, and at the same time determine the positions on the substrate P to expose the patterns B～D, so as to The patterns A to D are bonded on the substrate P and the screens are merged into a drawing board pattern.

Utilizing this picture combination, each pattern A to D is all just overlapped with the mark MK formed on the zeroth layer on the substrate P with high accuracy, and each of the patterns A to D is overlapped with the zeroth layer with high accuracy. , as a result, the bonding accuracy of the patterns A to D can be improved.

Then, the pattern (source/drain pattern, etc.) exposed after the second layer on the substrate P is aligned with the substrate formed by the gate pattern of the first layer during exposure when replacing the grating with the pattern. mark, so as to align the grating and adjust the imaging characteristics of the projection optical system 12, and superimpose on the grid pattern with high accuracy. In addition, in order to overlay the second layer and subsequent layers on the first layer with high accuracy, for other methods using substrate alignment marks, the above-mentioned test exposure can be performed on the exposure patterns of the second layer and subsequent layers, and pre-bonding can be used. This pattern is used to obtain the correction parameters at the time of exposure.

In the exposure method and exposure device of the above-mentioned embodiments of the present invention, because in the pre-exposure process, the transferred gate pattern and mark MK are firstly exposed on the substrate P by a program, and then calculated in advance according to the overlap deviation amount. Correction parameters when exposing the gate pattern, so in the case of exposing the gate pattern that exists on the zeroth layer and overlaps the zeroth layer on the substrate P, exposure with the same accuracy can be performed. As a result, the When bonding a plurality of patterns on the zeroth layer, the bonding accuracy can be sufficiently improved.

Furthermore, in this embodiment, two-dimensional marks are used as the marks MK, and bonding errors in both the X direction and the Y direction can be detected from each mark MK exposing the bonding pattern. Therefore, in the pre-exposure process, not only the correction parameters for each pattern are obtained, but also the relative bonding accuracy between the bonded patterns can be obtained. Therefore, in this embodiment, the use of correction parameters not only pursues the alignment accuracy of each pattern A to D, but also improves the joining accuracy. For example, using the pattern A as a reference, using the relative joining of the corresponding pattern A Correction parameters of accuracy correction patterns B to D are joined together.

However, in this embodiment, since the gate pattern is a positive film pattern and the mark MK is a negative film pattern, at the same time, the arrangement pitch of the scanning line pattern 38 and the gate electrode pattern 39 corresponds to the arrangement pitch of the mark MK for overlapping exposure, so the amount of overlap deviation is measured The time range can be narrowed down to tens of μm. Therefore, when the measurement magnification is increased in the measuring device, edge detection can be performed without affecting the deterioration of measurement accuracy caused by the magnification of the objective lens, and high-accuracy detection can be performed. It is possible to contribute to high accuracy and cost reduction by using overlapping measuring devices and the like without additionally supplying new measuring devices. In addition, in this embodiment, since the mark MK on the grating RT is arranged near the center of the grating, the error of the grating components caused by the magnification or rotation during projection is minimized, and high-precision bonding can be realized. .

Moreover, in the above-mentioned embodiment, the mark MK as the first mark is formed on a grating RT different from the grating R having a gate pattern. Of course, this is not intended to limit the present invention. A first mark is formed on the raster R of the pattern. In this case, the first mark is arranged at a corner part away from the pattern A, and the test exposure is performed with the first mark positioned near the optical axis of the optical projection system 12 by using the movable grating stage 13, so that Errors in raster components caused by magnification or rotation during projection are minimized.

In addition, in the above-mentioned embodiment, the grid pattern is made of a positive pattern, and the mark MK as the first mark is made of a negative pattern. On the contrary, of course, the grid pattern can also be made of a negative pattern. The mark MK can also be made of a positive pattern. Moreover, the pattern of the first mark is not limited to the above-mentioned shape, and it may be a mark MK in which the edges in the X direction and the Y direction are L-shaped, as shown in FIG. 7 .

In addition, in the above-mentioned embodiment, the pre-exposure process can expose and form four marks MK on the substrate P in one exposure. Of course, as shown in FIG. The pattern is bonded while the individual exposures form 2 marks MK. Moreover, in the above-mentioned embodiment, the patterns A to D to be bonded are described as patterns having the same gate pattern. Of course, the present invention is not limited thereto, and can also be applied to the case of bonding different patterns.

In addition, the substrate of this embodiment is not limited to glass plates for liquid crystal display elements, and can also be applied to semiconductor wafers for semiconductor elements, ceramic wafers for thin-film magnetic heads, or original plates for masks or gratings of exposure devices. (synthetic quartz, silicon wafers).

As far as the exposure device 9 is concerned, it is possible to use a step and repeat (Step And Repeat) exposure device (stepper) in which the pattern of the grating R is exposed when the grating R and the substrate P are in a stationary state, and the substrate P is moved step by step. Stepper) or other scanning exposure devices (Scanning Stepper, Scanning Stepper; USP5,473,410) can also be applied by moving the grating R and the substrate P at the same time to scan and expose the step and scan (Step AndScan) pattern of the grating R pattern.

As far as the type of exposure device 9 is concerned, it is not limited to the exposure device for the manufacture of a liquid crystal display element that exposes a pattern of a liquid crystal display element on the substrate P, and can also be widely used for semiconductors that expose a pattern of a semiconductor element on a wafer. Exposure equipment for device manufacturing, or exposure equipment for thin-film magnetic heads, imaging devices (CCD), and gratings.

Moreover, as the light source of the beam B, not only rays (g-line (436nm), h-line (404.7nm), i-line (365nm), krypton fluoride (KrF) excimer laser) generated from an ultra-high pressure mercury lamp can be used. (Excimer Laser), (436nm), argon fluoride (ArF) excimer laser line (436nm), fluorine (F ₂ ) excimer laser line (436nm)), charged particle beams such as X-rays or electron beams can also be used . For example, when electron beams are used, thermionic discharge type lanthanum hexabodide (LaB ₆ ) or tantalum (Ta) can be used as the electron gun. In addition, when electron beams are used, a pattern of the grating R can be used, and of course a pattern in which the grating R is directly formed on the glass substrate can also be used. In addition, high-frequency radiation such as YAG laser or semiconductor laser may be used.

The magnification of the projection optical system 12 is not limited to an equal magnification system, and of course it may be one of a reduction system or an enlargement system. For the optical projection system, in the case of using far ultraviolet rays such as excimer laser lines, far ultraviolet penetrating materials such as quartz or fluorite as nitrate materials can be used; in the case of using fluorine laser lines or X-rays, it can An optical system using a catadioptric system or a refractive system (the grating R can also use a reflective type grating); or in the case of using an electron beam, an electron optical system consisting of an electron lens and a deflector as an optical system can be used . Furthermore, the optical path through which electron beams pass is of course in a vacuum state. In addition, of course, it is also applicable to a proximity (proximity) exposure apparatus which exposes the pattern of the grating R by contacting the grating R and the board|substrate P, without using the projection optical system 12. In addition, in the above-mentioned embodiments, the projection optical system is represented by a single lens, of course, a plurality of projection lenses can also be arranged so that the projection areas overlap each other, that is, a projection optical system with multiple lenses.

In the case of using a linear motor (LinearMotor) (please refer to USP5,623,853 or USP5,528,118) in the substrate stage 14 or grating stage 13, of course, the air floating type using the air bearing (Air Bearing) and the use of the Luo Magnetic floating type with Lorentz Force or Reactance force. And, each carrying platform 13,14 can be the form along guide rail, or also can be the guide lens form (Guide Lens Type) that guide rail is not provided.

The driving mechanism of each stage 13, 14 can be driven by an electromagnetic force that makes a magnet unit (permanent magnet) in which magnets are arranged in two dimensions and a motor subunit in which coils (coil) are arranged in two dimensions to face each other. Planar motors for each of the carrying platforms 13, 14. In this case, either one of the magnet unit and the motor subunit contacts the stage 13 , 14 , and the other of the magnet unit and the motor subunit is provided on the moving surface side (the base) of the stage 13 , 14 .

The reaction force generated by the movement of the substrate stage 14 is not transmitted to the optical projection system 12, as described in Japanese Patent Application Laid-Open No. 8-166475 (USP5,528,118), which uses a frame (Frame) member and mechanically released into the bed (substrate). Therefore, the present invention can be applied to an exposure apparatus having such a structure.

The reaction force generated by the movement of the grating carrier 13 is not introduced into the optical projection system 12, as explained in Japanese Patent Laid-Open No. 8-330224 (US S/N 08/416,558), which uses a frame The components are mechanically released into the bed (substrate). Therefore, the present invention can be applied to an exposure apparatus having such a structure.

As mentioned above, the exposure device 9 as the substrate processing device of this embodiment is a variety of components that can ensure the mechanical accuracy, electrical accuracy, optical accuracy, etc. of the settings, and includes various components. Subsystem (Sub system) assembled and manufactured. In order to ensure the above-mentioned various accuracies, before and after assembly, it is necessary to perform adjustments to achieve optical accuracy of each optical system, adjustments to achieve mechanical accuracy of each mechanical system, and adjustments to achieve electrical accuracy of each electrical system. The step of assembling various subsystems into an exposure device includes mechanical connections among various subsystems, wire connection of circuits, conduit connection of pneumatic circuits, and the like. Before assembling the various sub-systems into an exposure apparatus, it goes without saying that individual assembling steps of each sub-system are also included. After assembling various sub-systems into an exposure device, an integrated adjustment is required to ensure various accuracies of the entire exposure device. In addition, the fabrication of the exposure device is preferably performed in a clean room where temperature and cleanliness can be managed.

As shown in FIG. 9, a liquid crystal display element or a semiconductor element includes a step 201 of designing the function and performance of the liquid crystal display element, etc., a step 202 of making a grating R (mask) based on the design step, and making a glass plate made of quartz or the like. Step 203 of P or a wafer made of silicon material, step 204 of exposing the grating R pattern on the glass plate P (or wafer) using the exposure device 9 of the above embodiment, step 205 of assembling liquid crystal display elements, etc. (in In the case of a wafer, it includes a dicing (Dicing) process, a welding (Bonding) process, a packaging (Package) process, etc.) and an inspection step 206 .

As described above, in the exposure method of the present invention, after exposing the bonding pattern and the first mark on the substrate, the correction amount for exposing the bonding pattern is sequentially set according to the relative positional relationship between the first mark and the pattern.

Therefore, in this exposure method, even when bonding a plurality of patterns on a substrate, it is possible to achieve a sufficient effect of improving bonding accuracy.

Regarding the exposure method of the present invention, it includes sequentially exposing the first mark on each bonding pattern.

Therefore, in this exposure method, since not only the correction parameters for each pattern are obtained, but also the relative bonding accuracy between the bonding patterns can be obtained, it is also possible to achieve the so-called relative bonding accuracy for the reference pattern. To correct the correction parameters of joining other patterns.

Regarding the exposure method of the present invention, it includes arranging the first marks corresponding to the arrangement pitch of the unit patterns.

Therefore, in this exposure method, it is possible to narrow the range when measuring the amount of overlap deviation between the first mark and the unit pattern to about several tens of μm, and it is possible to perform edge detection when the detection magnification is increased in the measuring device, and it is not necessary to The deterioration of the measurement accuracy caused by the magnification of the objective lens can be affected, and the effect of so-called high-accuracy detection can be achieved.

Regarding the exposure method of the present invention, it includes sequentially overlapping and exposing the first mark and the pattern.

Therefore, in this exposure method, it is possible to narrow the range when measuring the amount of overlap deviation between the first mark and the unit pattern to about several tens of μm, and it is possible to perform edge detection when the detection magnification is increased in the measuring device, and it is not necessary to The deterioration of measurement accuracy due to the magnification of the objective lens can be affected, and the effect of enabling high-accuracy detection can be achieved.

Regarding the exposure method of the present invention, wherein the pattern is one of the negative pattern or the positive pattern, and the first mark is the other of the negative pattern or the positive pattern.

Therefore, in this exposure method, it is possible to narrow the range when measuring the amount of overlap deviation between the first mark and the unit pattern to about several tens of μm, and it is possible to perform edge detection when the detection magnification is increased in the measuring device, and it is not necessary to The deterioration of measurement accuracy due to the magnification of the objective lens can be affected, and the effect of enabling high-accuracy detection can be achieved.

Regarding the exposure method of the present invention, the patterns are sequentially formed on the first layer of the substrate.

Therefore, in this exposure method, even in the case where no reference pattern is formed on the substrate, it is possible to achieve a so-called sufficient effect of improving bonding accuracy.

Regarding the exposure method of the present invention, wherein the first mark is formed on the central pattern of the mask.

Therefore, in this exposure method, errors in grating components due to magnification or rotation at the time of projection can be minimized, and an effect of enabling high-precision splicing can be achieved.

Regarding the exposure method of the present invention, the bonding pattern is composed of the same pattern.

Therefore, in this exposure method, even when bonding a plurality of patterns on a substrate, it is possible to achieve a sufficient effect of improving bonding accuracy.

The exposure device of the present invention includes a device capable of storing a correction amount when exposing a bonding pattern according to the relative positional relationship between the first mark and the pattern, and bonding the pattern on the substrate according to the stored correction amount.

Therefore, in this exposure apparatus, even when bonding a plurality of patterns on a substrate, it is possible to achieve a sufficient effect of improving bonding accuracy.

Claims (9)

1. an exposure method is applicable to engage a pattern and the desired pattern that exposes on a substrate, and it is characterized by: this method comprises:

Carry out a preparation exposure technology and engage pattern with one with one first mark different that on this substrate, expose to this pattern; And

Carry out one and set technology, this settings technology is set the revisal amount when exposing this joint pattern according to this first mark and the relative position relation between this pattern of being exposed on this substrate.

2. exposure method as claimed in claim 1 is characterized by: comprising this first mark that exposes on each this joint pattern.

3. exposure method as claimed in claim 2 is characterized by: wherein this pattern has the unit cell pattern with a setting spacing arrangement; And

This first mark is arranged setting spacing.

4. exposure method as claimed in claim 1 is characterized by: comprising this first mark of overlapping exposure and this pattern.

5. exposure method as claimed in claim 1 is characterized by: wherein this pattern is that then this first is labeled as a positive pattern to a negative pattern; This first be labeled as this negative pattern then this pattern be this positive pattern.

6. exposure method as claimed in claim 1 is characterized by: wherein be formed with the pattern of plural layer on this substrate, and this pattern is the pattern that is formed at ground floor.

7. exposure method as claimed in claim 1 is characterized by: wherein this first mark is formed at the central authorities of cover curtain.

8. exposure method as claimed in claim 1 is characterized by: wherein this joint pattern is made of identical patterns.

9. an exposure device is applicable to engage a pattern to expose to desired pattern on a substrate, and it is characterized by: this device comprises:

One memory storage, this memory storage can engage relative position relation between the pattern according to one first mark different with this pattern that is exposed on this substrate with one, this revisal amount when engaging pattern of storage exposure; And

One compensating device, this compensating device can engage this pattern according to the revisal amount that is stored in this memory storage on this substrate.

Images