JPS586128A - Method and apparatus for correcting defect of photo-mask - Google Patents

Abstract

PURPOSE:To correct the defect of the photo-mask in high quality by detecting the size of a laser irradiating region and mounting a mechanism which can set a condition optimum for the size. CONSTITUTION:A metallic complex film 4 is formed onto a substrate 1 so as to coat the defect 3 generated to mask patterns 2 shaped by chromium, chromium oxide or these multilayers. Laser beams are irradiated to the film 4, and the precipitated film 6 of a metal, a metal oxide or the mixture is precipitated. A metallic complex of a section not precipitated is removed. Here, a solution obtaind by adding silver nitrate, carboxylic acid and an additive to an organic solvent is used as a metallic complex material. Silver carboxylic acid salt among the silver salt excellently dissolves into the organic solvent, and forms a superior film even when the organic solvent is evaporated.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method and apparatus for correcting white spot defects (missing defects M) on a photomask.

Among the defects that occur on photomasks, black spot defects (residual defects 1)
Regarding iii), correction by laser
9508), a significant process reduction has been achieved. On the other hand, the lift-off method is used for missing defects, that is, defects in which a part of a normal notch is missing. The 7-off method is (1) a step of applying a positive-reducing resist to the entire surface of a photomask having missing defects.

(2) A process of exposing only the missing defective part using a partial exposure method; (5) A step of exposing only the missing defective part Kj! by the exposure process. (4) Step of forming a metal film on the missing defect portion and its surroundings or the entire surface of the photomask using vacuum evaporation technology; (5) Removing the resist and at the same time removing the metal film formed on the resist. It has the disadvantage that it requires a large number of steps and time to remove it.

A good method using partial vapor deposition using seeds has also been proposed, but there are many practical problems such as the strength of the fine f# film and the uniformity of the film.

As a method for repairing missing defects that eliminates these defects and forms a highly accurate and uniform film in less than 1 x 9 hours, a metal complex is applied to the entire surface of the photomask having missing defects or only to the missing defect areas and their surrounding areas. A method has been proposed in which the defect is corrected by depositing metal by irradiating only the missing defect with a laser beam. The present invention relates to an improvement in a method for repairing missing defects using this metal complex.

That is, an object of the present invention is to eliminate the drawbacks of the prior art described above and to provide a method and apparatus for correcting photomask defects with high quality.

By the way, when a coated metal complex salt is irradiated with sheath light to precipitate metal to correct a defect (a defect where a part of the normal pattern is missing), the size of the sheath irradiation area, that is, the size of the defect) K The more optimal sheath irradiation power density is different.

Therefore, the present invention provides a mechanism for detecting the size of the laser irradiation area and setting the optimum conditions according to the size, thereby achieving high-quality defect correction.

The present invention will be explained in detail below with reference to the drawings. First, FIG. 1 shows a factory in which defects are corrected by irradiating a metal complex with a laser to precipitate a metal oxide.

(a) A metal complex is coated on a glass substrate 1 so as to cover the missing defective portions that occur in the mask pattern 2 formed of chromium oxide, chromium oxide, or a multilayer thereof, and dried to form a metal complex film 4.

<h) Metal complex in the region including at least the missing defect part.

The body membrane 4 is irradiated with a laser beam 5, and a deposited film 6 of metal, metal oxide, or a mixture thereof is formed on the area irradiated with the laser beam 5.
is precipitated.

(C) After laser irradiation, remove the unprecipitated metal complex. Thereafter, cleaning, heat treatment, etc. are performed as necessary.

The metal complex materials used here include silver nitrate and organic solvent.
Nine solutions are used to add carboxylic acid and additives. At this time, carboxylic acid silver salt is formed in a dissolved state in the solution. Among the silver salts, carboxylic acid silver salts are preferably used by dissolving them in an organic solvent and evaporating the organic solvent.
This is to form a good film. Although various carboxylic acids can be used, among dicarboxylic acids having two carboxylic acid groups, citraconic acid has particularly excellent film-forming properties. It was organic +1#,! In addition to organic solvents such as alcohol-based, Seven-day Solp-based, carpitol-based, and glycol-based solvents, dimethyl sulfoxide, dimethyl acetoand, dimethyl formamide, etc. can be used; however, a mixture of methyl cellosolve, acetonitrile, acetylacetone, and dimethyl sulfoxide can be used. Liquid is best.

The objective can be further achieved by adding KFi titanium alcoholate which improves the adhesion between the precipitated metal J [or metal oxide film or its mixture film and the substrate], but titanium tetrabutylide is most suitable. F) 1!11 is applied as a coating liquid and dried to form a complex film of 0.1 to 1 μm, and a predetermined portion is irradiated with radar. After a metal oxide film or a mixture thereof is deposited by laser irradiation, methyl cellosolve is washed with acetonitrile or a mixture of both to completely remove the complex film other than the precipitated part. can be removed. Thereafter, baking in an ultraviolet oven for 30 seconds to 5 minutes can improve the strength of the deposited film and the strength of adhesion to the substrate.

The irradiation is performed using a sea oscillator (see the fs2 diagram).
(not shown) is formed into a rectangular shape by a variable slit 7 that can change the size to any size, and an objective lens 8 places the laser beam 5 at a position where the actual size of the variable slit 7 connects. An objective lens 80 is placed on a photomask 10 having a
The second beam is focused into a rectangular shape with the size of the reciprocal of the magnification.
The optical system shown in the figure allows the complex film to be spread approximately (1.21 m
The tsS diagram shows the results of depositing on a sample formed to a thickness of 200 nm by changing the power density of the irradiation area and the power density at the irradiation part. In the figure, the area above the solid line is where the power density is too high and the deposited film will be damaged by the laser, and the area below the broken line is where the power density is too low and there is a high risk that the deposited film will fall off during cleaning. It is. In other words, the solid line and the broken line are 8★
The area covered by this is the appropriate condition range.

The conditions shown by the one-dot chain line were optimal. The horizontal axis is the size of the irradiation area (the length of one side when irradiating a square), and the vertical axis is the rape power level of the irradiation area. In the range of the irradiation area size (5) from 2 μm to sosw*, the optimum condition was that the radar power severity ψ) satisfied the 5xp-constant relationship.

Figure 5 shows the results when a square area is irradiated, but when the irradiation area is rectangular, the optimum power density is when the square root of the area of the irradiation area is taken as the dimension (5). Ta. When the nine-dashi irradiation area had a rectangular shape with a ratio of short side to long side of 1:5 or more, it was better to divide the area into several areas and irradiate the skin.

Figure 5 shows organic solution IK nitric acid #1. These are the results when a solution containing carboxylic acid and additives is used to form a film with a thickness of 1.5 μm, and the appropriate power density varies depending on the composition of the complex and the film thickness.

Also silver nitrate in an organic solvent as a metal complex solution.

The present method is not limited thereto, but rather refers to metal complexes in the narrow sense, i.e., metal complexes with one or more metals as the central atom and other metal complexes. Atom f& is an atomic group, and in addition to the atomic group when diagonal ligands combine to form the atomic group 1'), it also applies to the concept including metal salts produced in normal solution reactions. Self-defense.

An example of a correction device for carrying out the correction method II described above will be shown.

The defect correction device of the embodiment shown in Fig. ls4 includes an Ay receiver oscillator 11 and a ninth tool 1 which generates 50 optical paths of non-receiving light.
2. Shutter 15, zoom optical system 14 and its driving device 5
1. A make-up mirror 15 that bends the laser beam toward the objective lens 8, a nine rectangular open loss slit 7 disposed on the optical path of the laser beam 5 toward the objective lens 8, and an O drive device 16 therefor.
and detection value R17 for detecting the slit position, No.-7 electric tool 18. Objective lens8. An illumination light source 19 for observing and aligning the mask pattern 2 of the photomask 10, a No. 7 mirror 20, an observation optical system 21, a laser cut filter 22, a slit illumination light source 25, and an interference filter 24. Pattern of the photomask 10. A half mirror 25 for observing and displaying the projection of the rectangular aperture slit 70
And Shirezu Cutler 4 Ruta 26 and Imaging Device 27 and Minejiri T
X-Y table 2 on which i2B, 7 Otoma Airi 10 is placed
9 and its drive device go, a detection device 17 that controls the drive device 50 of the X-Y table 29 and the drive device 15, and also controls the drive device 16 of the rectangular opening slit 7 to detect the dimensions of the rectangular opening at that time. The control device 52 determines the dimensions of the laser irradiation area based on signals from the laser beam, and controls the drive device of the x-luminescent optical system 14 to control the beam diameter to obtain the optimum paris density. .

The radar light 5 oscillated from the JR Schiller oscillator 11 is bent in its optical path by a make-up mirror 15, and reaches a rectangular open loss filter 7 whose width in both X-1 directions can be set independently. - It is formed into a rectangular shape of a.

The laser beam 5 formed into an arbitrary rectangular shape by the six inter-rectangular slits 7 passes through the No-Funk 25.18, is focused by the objective lens 8, and is irradiated onto the photomask IOK having the missing defect. In ζζ, the photomask 10 and the rectangular aperture slit 7 are reduced by the objective lens 6 to the reciprocal of the 80 magnification of the objective lens, that is, when the 80 magnification of the objective lens is 1, /HK is reduced. placed in the projected position.

Observation and alignment of the pattern 2 of the photomask 10 can be performed by the observation optical system 21 by combining the light from the illumination light source 19 with the /S-function 2018. Further, a shading light cut filter 22 is arranged in the observation optical system 21 for the safety of the operator.

Furthermore, the light from the slit illumination light source 25 is filtered through an interference filter 24.
This allows only specific wavelengths to pass through, making it extremely difficult to make waves.
15 and is projected onto the photomass 10 by the objective lens 8 as a projection through the rectangular opening slit 70. Here, the light beam 1)-15 has a characteristic of reflecting only the wavelength of the sheath light 5 and transmitting a specific wavelength of the slit illumination light.

The image of the rectangular opening slit 7 created by this slit illumination light can also be observed using the observation optical system.

Further Maku, Halfumi 9-25. Rays cut filter 26. and imaging device di! Photomask 10 by 27
It is also possible to observe the projected image of pattern 2 and rectangular opening slit 7 on the monitor I〆28.
A metal complex solution is applied onto the photomask 10 having the above, and a metal complex JI[4 is formed by drying.

The missing defect portion 55 can be adjusted using the X-Y table 29 so that it coincides with the projection of the rectangular slit 70. Further, the X-Y table 29 can be driven manually or by a drive device 50.

The observation optical system 21 adjusts the width of the rectangular aperture slit 7 while observing with the peaked TF 28 to make the projected image match the size of the missing defect portion 5M. At this time, the slit position detection device 17 detects the inside of the slit in both the X and Y directions,
The control device 52 calculates the optimum power density by the width KTh, and drives the drive device 31 of the zoom optical system 14 to set the beam diameter to obtain the optimum power density.

When all the settings are completed, the shutter 1s is opened and the sheath light 5 is irradiated to the missing defective part SSK. At this time, the laser beam 5 is irradiated onto exactly the same area as the projected image of the rectangular opening slit 70 by the slit illumination light.

Here, the shutter 15 is used to turn on and off the sheath light 5, but the electro-optic trout is modulated by the acousto-optic effect! The same operation can also be performed by controlling the power supply of the KAr Lacey oscillator 11.

The control device 52 has a gate function. That is, the position information detected by the defect inspection device is read from a magnetic medium or the like, and the rectangular opening slit 7 is used to drive the X-Y table 29 and receive signals from the slit position detection device 17.
It is necessary to have a function to calculate the 0 dimension (that is, I times 2 M of the rectangular shape projected on the photomask 10 is the magnification of the objective lens) and calculate the optimal power density from that dimension, and to obtain the optimal power density. By calculating the beam diameter and adjusting the beam diameter to a predetermined value by driving the drive device 51 of the zoom optical system 14, and by controlling the shutter 1iTo, modulator, and power supply under appropriate conditions, the laser beam 5 is ON, O
It has a function of FF. If the width of the missing defective portion S3 is known in advance, it is also possible to set the width of the rectangular opening slit 7 by driving the drive device 1i271 for changing the width of the rectangular opening slit 7.

Here, the proper conditions for turning on and off the laser beam 5 are, for example, detecting the amount of light transmitted through the photomask 10 of the irradiated laser beam 5, and determining the state of precipitation of metal etc. from the change in the amount of transmitted light. t containing the conditions obtained.

Next, one aspect of a method for correcting the missing defect A5 of the photomask 10 using the defect correction apparatus shown in the forty-first embodiment will be described. Photomask 10 with missing defect s5'6
The entire i is a missing defect! After applying a metal complex solution such as a silver complex solution to the black ink and its surrounding area 11 and drying it to form a metal complex film 4, it is placed on an X-Y table 29, and based on the position information from the defect inspection device. Control-place! 2 drives the table driving device 50 and at least the ttas optical system 21.
Alternatively, the missing defective portion may be placed within the field of view of the imaging device 27. Next, while observing the projected image of the rectangular opening slit 70 by the slit illumination light source 2iSK, positioning is performed using the x-y table 29 and the slit driving device 16 is driven to set the inside of the slit. At this time, the rectangular opening slit 7
Adjustment is made so that the projected image does not protrude from the normal pattern and sufficiently covers the missing defective portion SS. Here, the slit position detection device 17 sends a signal that presses the distance from the optical axis (the center of the optical system) to the slit tip position to the control device s2, and sends two signals in the X direction and Y The dimensions of the rectangular opening are calculated from the two direction signals. Then, in the control device s2, the optimum power density for these dimensions and dimensions is determined by judging from an approximate formula or a table stored in the memory section. That is, in the example shown in Fig. 5, since there is a certain relationship PxSw between the optimum power density P and the slit size (dimension on the projection plane) S, the power density at 10μ swearing is 400oWA. If you set it to
That is, it is sufficient to drive the zoom drive device si to widen the beam diameter by f times. (The power density is inversely proportional to the square of the beam diameter.) After all settings are completed, the shutter 15 is opened automatically or according to the operator's instructions, and the Ar laser oscillator 1 is opened.
A laser beam 5 oscillated from 1 is irradiated onto the missing defect portion s3. Next, after irradiating for a certain period of time or while monitoring the state of precipitation of metal, etc., a signal is sent to the shutter 15 under appropriate conditions to stop the irradiation of the laser beam 5.

As described above, the metal complex l[4 is irradiated with laser light 5,
By depositing the metal film, the missing defect 55 of the photomask 10 is repaired.

This completes the correction of - number of missing defects 3s, and after observing if necessary, when the operator issues a signal indicating completion of correction, the X-Y table 29 is driven to the next defect position. After the metal film has been deposited on all the missing defective parts 35, the photomask 10 is cleaned to remove the gold film from the parts other than the parts where the metal has been deposited.
After removing the I4 integral film 4 and applying heat treatment if necessary, the repair work is completed.

FIG. 5 shows another example of the photomask missing defect correction apparatus. The correction device shown in FIG. 5 differs from the correction device shown in FIG. 4 in the following points. That is, an aperture plate 35 having a predetermined size and a light amount detector 56 are provided on the back side of the dichroic minnow S4. The dichroic mirror 54 has a characteristic of reflecting most of the laser beam 5, transmitting a portion, for example, the first beam, and transmitting the light from the slit illumination light source 25 selected at a specific wavelength by the interference filter 24. Only the center portion of the beam transmitted through the dichroic mirror 34 passes through the beam B5, which has an aperture having a certain area, and is received by a photodetector 561ft. Here, the transmittance of Guikuroitsukumi 2-54 to the screen light 5 and that (-on the slit 7 surface if the area of 35 is known)
(Word density, that is, on the photomask 10) (The word density can be calculated. From this, the slit size of the rectangular opening slit 7 is detected, and the control device 32 calculates the optimum word density for that size) using an approximate formula or memory. After determining the value from a memorized table, the zoom optical system 14 is driven by the zoom drive device 51 to change the beam diameter while comparing it with the signal from the light amount detector 36, and the signal from the light amount detector 56 is changed. It is sufficient to stop the zoom drive device s1 when the optimum power density matches or falls within a certain error (for example, ±5%).As described above, the irradiation power density of the laser beam 5 on the photomask 10 can be set to an optimal value.

Next, the zoom optical system 14 used in the correction device shown in FIGS. 4 and 5 will be explained.

Figure W46 shows an example of the zoom optical system 14, which includes a concave lens 58, a convex lens 59. Concave lens 40. and convex lens 4
1, which constitutes an afocal zoom system (when parallel light is input, it is output as parallel light).
Ru.

When the convex lens 39 and the concave lens 40 are combined as shown in FIG. 6(A)K, the incident filtered light 5 is output with the maximum beam diameter. On the other hand, as shown in (@)k, the convex lens 3
9 and concave lens 40 are respectively concave lens 58 and convex lens 4.
1, output with the minimum beam diameter. The convex lens 69 and the concave lens 40 are shown in FIG.
) By continuously changing the hole position shown in
The output beam diameter changes continuously. Of course, the beam diameter can also be reduced by selecting the focal length of each lens.

In addition to the above-mentioned afocal zoom system, the same function can be obtained by combining, for example, a camera zoom lens and a condensing lens with a fixed focal length. .

Furthermore, in the correction apparatus shown in FIGS. 4 and 5, we have explained the case where the zoom optical system 14 is used as a means for changing the power density, but as shown in FIG. chromium, aluminum, nichrome,
A filter in which the transmittance is continuously changed by a thin film formed by vapor deposition or the like using a metal such as Romel with varying thickness in the local direction, or a transparent flat plate 44°44' perpendicular to the optical axis as shown in Figure 8. installed so that it is plane symmetrical with respect to the plane 45,
The incident angle of the laser beam 5 changes continuously (however, it maintains a plane symmetrical relationship with respect to the plane 45 perpendicular to the optical axis, 7'Ett
) By using a filter whose transmittance changes continuously by rotating it in the direction shown by the K arrow,
The purpose will be achieved.

It is also clear that a similar function can be obtained by using the zoom optical system 14 and the filter shown in FIG. 7 or 8 in combination. When using the filters shown in FIGS. 7 and 8, measures such as enlarging the beam diameter must be taken in advance to obtain a sufficiently uniform beam power distribution up to the maximum area irradiated with the laser beam 5. It is necessary to keep it.

Continuous wave laser light is preferable as a source for carrying out the present invention, and although the case where an Ay receiver oscillator is used has been explained in the embodiments, laser light with near-infrared, visible, or ultraviolet wavelengths may be used. Good. This includes the fundamental wave of YAG laser, 1s2 harmonic and third harmonic, Kr laser,
Ih-Cd lasers, continuous wave continuous pumping Dya lasers, and 4r lasers are included, but it is preferable to use short visible wavelengths to near ultraviolet wavelengths.

As explained above, the present invention requires fewer steps than the conventional 9-ft-off method. It is possible to repair the missing defect in a time of 100 minutes, and the adhesion strength between the metal film, metal oxide film, or a mixture thereof deposited on the missing defect and the substrate is high, and the optical property can be sufficiently repaired with high precision. can be done,
Therefore, the production yield of semiconductor devices can be greatly improved.

Furthermore, the apparatus of the present invention has the effect of reliably implementing the missing defect correction method.

[Brief explanation of the drawing]

Fig. 1 is a diagram showing the steps of the method of the present invention, and Fig. 2 is an explanatory diagram of the optical system for carrying out the method of the present invention. Diagram showing the relationship between dimensions and optimum density, #! FIG. 4 is a system diagram showing an example of a missing defect correction apparatus for carrying out the method of the present invention. FIG. 5 is a system diagram showing another example of a missing defect correction apparatus for carrying out the method of the present invention. I (work density tL
@A diagram showing an example of a zoom optical system as an adjusting means, FIG. 7 and an sIs diagram are the same) (A diagram explaining a continuously variable transmittance filter as a word density adjusting means. 5... Laser light, 7 ...Rectangular opening slit 8
...Objective lens, 10...Photomask, 11...
・Ar oscillator, 13...Shirt oscillator, 14...
-Zoom optical system. 15... Guicroic mirror. 29... X-Y table, 17... Slit position detection device, 52... Control device, 36... Light amount detection device. Sai 1 Mouth 3 WJ 1 S (μm) Oki Figure 1 ( 5 Figure O 6 CB > Off Trap 3 8 Years Old, 44 Continuation of Page 1 0 Inventor Ryōaki Okunaka City of Yokohama 292 Yoshida-cho, Totsuka-ku Hitachi, Ltd. Production Technology Laboratory C0 Author: Takao Yonabe 0 Author: Isao Tanabe 1450 Josui Honcho, Kodaira-shi Musashi Factory, Hitachi, Ltd.

Claims (1)

[Claims] 1. After applying the wnt* group complex S* to at least four missing defect areas and their surroundings of a photomask having missing defects,
Bar 9- according to the size of the laser irradiation area including the missing defect part
V! A method for repairing missing defects in a photomask, characterized in that the metal film and metal oxide film tx are deposited as a mixture film by irradiating the laser irradiation value range with nine-sea light set at an R degree. 2. Correction of missing defects in a photomask as described in item 1 of the scope of the patent application, characterized in that the metal complex #liquid is made by adding at least silver nitrate, carboxylic acid, and titanium alcoholate to an organic #rope. Method. A means for generating sheath light, a rectangular aperture slit for shaping the laser light generated by the scissor means into a rectangle of arbitrary size, and a sheath light formed into a rectangle by the rectangular aperture slit. an objective lens for condensing and forming a reduced image of the image, and a photo disk having a missing defect; 6. an X-Y table, a means for detecting the dimensions of the rectangular aperture slit, and a means for arbitrarily changing the power density of a laser beam condensed and imaged as a reduced image of the slit; means for making the optical beam os, oipp; and a reduced image of the bell-shaped Ross slit based on information on the defect position. The size of the slit is determined based on the function of controlling the position of the X-Y table and the information from the means for detecting the size of the rectangular opening slit,
A function to calculate the optimum power density for the dimensions, a function to control the means for changing the power density of the laser beam to obtain the calculated optimum power fMR, and a function to control the means to opi and oyp the sheath light. What is claimed is: 9. A photomask chipping/defect correction apparatus, comprising: a control device having a control device; 4. A means for generating a laser beam, a rectangular aperture slit for shaping the laser beam generated by the mosquito means into an arbitrary rectangle, and a rectangular aperture slit for shaping the laser beam into a rectangular shape by the rectangular aperture slit. An objective lens for condensing light as a small image, an X-Y table on which a photomask having a missing defect is placed, means for detecting the dimensions of the rectangular aperture slit, and a reduced image of the rectangular aperture slit. means for arbitrarily changing the power density of the laser beam condensed and condensed as a laser beam; means for detecting the changed power density of the sheath beam; means for turning the laser beam ON and OFF; Based on the information, the reduced image of the rectangular aperture slit matches the missing defect position K. The size of the slit is determined by the function of controlling the position of the The function of calculating the optimum power density for the dimensions and the information from the means for detecting the power density are compared with the calculated optimum power density, and the laser beam is adjusted so that the two match or fall within a certain error range. What is claimed is: 1. A photomask missing defect repairing device comprising: a control device having a function of controlling a means for changing para-density and a function of controlling a means for turning on and off a laser beam. 5. The control device stores the relationship between the dimensions of the rectangular opening slit and the optimum power density as an approximate expression, and determines the optimum power density that falls within the slit dimensions determined based on information from the means for detecting the dimensions of the rectangular opening slit. The approximation formula or mask missing defect correction device. 6. The control device stores the relationship between the dimensions of the rectangular opening slit and the optimum power density as a table, and calculates the optimum power for the slit dimension determined based on information from the means for detecting the dimensions of the rectangular opening slit. 5. The photomask missing defect correcting apparatus according to claim 3, further comprising a function of determining the density by referring to the table. '7. Device for correcting missing defects in the power density of the laser beam. 8. 4th JJ
The apparatus for correcting missing defects in a photomask according to I. 9. Continuous oscillation lr as means for generating the laser beam
An apparatus for correcting missing photo disk defects as set forth in claim 3 and claim 4, characterized in that it is equipped with a receipt scratcher.