Mask for manufacturing semiconductor device and method of manufacture thereof

What is claimed is:

1. A mask for manufacturing a semiconductor device comprising:

a transparent substrate having main and back surfaces;

a circuit pattern formed on said main surface and comprising a first region comprising a metal layer and a second region where said transparent substrate is exposed; and

a means for shifting a phase of light, formed at both ends of said first region and in a part of said second region and not in the whole of said second region.

2. A mask for manufacturing a semiconductor device according to claim 1, wherein said means for shifting phase of light is formed in contact with at least a portion of said first region which is in the vicinity of said second region.

3. A mask for manufacturing a semiconductor device according to claim 2, wherein said means for shifting phase of light is a transparent film of a material selected from the group consisting of indium oxide, silicon dioxide, silicon nitride, magnesium fluoride, and polymethyl methacrylate.

4. A mask for manufacturing a semiconductor device according to claim 2, wherein said means for shifting phase of light is a groove formed extending into said transparent substrate.

5. A mask for manufacturing a semiconductor device comprising:

a transparent substrate having main and back surfaces;

a circuit pattern formed on said main surface and comprising a light shield region comprising a metal layer and a transmission region wherein said transparent substrate is exposed, the lower surface of said metal layer being in contact with said transparent substrate;

a second groove in a part of the light shield region and having a depth reaching said transparent substrate from an upper surface opposite to a lower surface, and said first groove for shifting phase of light being located below said second groove.

6. A mask for manufacturing a semiconductor device according to claim 5, wherein said means for shifting phase of light is a transparent film of a material selected from the group consisting of indium oxide, silicon dioxide, silicon nitride, magnesium fluoride, and polymethyl methacrylate.

7. A mask for manufacturing a semiconductor device according to claim 5, wherein said means for shifting phase of light is a groove formed extending into said transparent substrate.

8. A mask for manufacturing a semiconductor device comprising:

a transparent substrate having main and back surfaces;

a circuit pattern formed on said main surface and comprising a first region comprising a light shield region for shielding irradiation light and a transmission region where said transparent substrate is exposed; and

a means for shifting phase of light, formed in a part of said transmission region and not in the whole thereof, wherein the means for shifting phase of light has a thickness x, the thickness x of said means for shifting phase of light is,

x=.lambda./(2(n.sub.1 -1))

.lambda.: wavelength of irradiation light

n.sub.1 : reference index of said means for shifting phase of light.

9. A mask for manufacturing a semiconductor device according to claim 8, wherein said means for shifting phase of light is formed in contact with at least a portion of said light shield region which is in the vicinity of the transmission region.

10. A mask for manufacturing a semiconductor device according to claim 9, wherein said means for shifting phase of light is a transparent film of a material selected from the group consisting of indium oxide, silicon dioxide, silicon nitride, magnesium fluoride, and polymethyl methacrylate.

11. A mask for manufacturing a semiconductor device according to claim 9, wherein said means for shifting phase of light is a groove formed extending into said transparent substrate.

12. A mask for manufacturing a semiconductor device according to claim 11, wherein the thickness x of the means for shifting phase of light is the depth d of said groove, the depth d being,

d=.lambda./(2(n.sub.2 -1))

.lambda.: wavelength of irradiation light

n.sub.2 : refractive index of the transparent substrate.

13. A mask for manufacturing a semiconductor device according to claim 8, wherein said means for shifting phase of light is formed in the central part of said transmission region.

14. A mask for manufacturing a semiconductor device according to claim 13, wherein said means for shifting phase of light is a transparent film of a material selected from the group consisting of indium oxide, silicon dioxide, silicon nitride, magnesium fluoride, and polymethyl methacrylate.

15. A mask for manufacturing a semiconductor device according to claim 13, wherein said means for shifting phase of light is a groove formed extending into said transparent substrate.

16. A mask for manufacturing a semiconductor device according to claim 14, wherein the thickness x of said means for shifting phase of light is,

x=.lambda./(2(n.sub.1 -1))

.lambda.: wavelength or irradiation light

n.sub.1 : refractive index of said means for shifting phase of light.

17. A mask for manufacturing a semiconductor device according to claim 15, wherein the depth d of said groove is,

d=.lambda./(2(n.sub.2 -1))

.lambda.: wavelength of irradiation light

n.sub.2 : refractive index of said transparent substrate.

18. A method of manufacturing a mask comprising:

a) preparing a transparent substrate having main and back surfaces;

b) forming a metal layer over said main surface;

c) coating a resist on an upper surface of said metal layer in a desired circuit pattern;

d) etching said metal layer to form said circuit pattern, leaving a part of the metal layer, so as to form a transmission region where said transparent substrate is exposed and a light shield region comprising said part of said metal layer; and

e) determining a region where a means for shifting phase of light is formed by magnifying said circuit pattern, and forming said means for shifting phase of light, in a part of said transmission region, said means for shifting phase of light being in said part of said transmission region and not in the whole thereof.

19. A method of manufacturing a mask according to claim 18, wherein said means for shifting phase of light is formed at both ends of said light shield region and in contact with said light shield region.

20. A method of manufacturing a mask according to claim 19, wherein said means for shifting phase of light is a transparent film of a material selected from the group consisting of indium oxide, silicon dioxide, silicon nitride, magnesium fluoride, and polymethyl methacrylate.

21. A method of manufacturing a mask according to claim 19, wherein said means for shifting phase of light is a groove formed extending into said transparent substrate.

22. A method of manufacturing a mask according to claim 18, wherein said means for shifting phase of light is formed at an area surrounding said light shield region and in a central part of said transmission region.

23. A method of manufacturing a mask according to claim 22, wherein said means for shifting phase of light is a transparent film of a material selected from the group consisting of indium oxide, silicon dioxide, silicon nitride, magnesium fluoride, or polymethyl methacrylate.

24. A method of manufacturing a mask according to claim 22, wherein said means for shifting phase of light is a groove formed extending into said transparent substrate.

25. A method of manufacturing a mask according to claim 18, wherein said means for shifting phase of light is formed by a focused ion beam.

26. A mask for manufacturing a semiconductor device according to claim 1, wherein said transparent substrate is a transparent glass substrate.

27. A mask for manufacturing a semiconductor device according to claim 8, wherein said transparent substrate is a transparent glass substrate.

28. A method of manufacturing a mask according to claim 18, wherein said transparent substrate is a transparent glass substrate.

29. A mask for manufacturing a semiconductor device according to claim 1, wherein said second region has corners, and wherein sub-transmission regions are provided at said corners so as to increase light intensity at the corners.

30. A method of manufacturing a mask according to claim 18, wherein the transmission region has corners, and wherein the method includes the further step of forming sub-transmission regions at the corners, so as to increase light intensity at the corners.

31. A method of manufacturing a semiconductor device according to claim 3, wherein said transparent film extends over said first region, and extends so as to formed in said part of said second region.

32. A mask for manufacturing a semiconductor device according to claim 4, wherein said groove is arranged along the contour of said metal layer.

33. A mask for manufacturing a semiconductor device comprising:

a transparent substrate having main and back surfaces;

a circuit pattern formed on said main surface and comprising a first region comprising a metal layer and a second region where said transparent substrate is exposed; and

a means for shifting a phase of light, formed in a part of said second region and not in the whole thereof, and adjacent to a periphery of said first region.

34. A mask for manufacturing a semiconductor device according to claim 33, wherein said transparent substrate is a transparent glass substrate.

35. A mask for manufacturing a semiconductor device according to claim 33, wherein the thickness x of said means for shifting phase of light is,

x=.lambda./(2(n.sub.1 -1))

.lambda.: wavelength of irradiation light

n.sub.1 : refractive index of said means for shifting phase of light.

36. A mask for manufacturing a semiconductor device according to claim 34, wherein said means for shifting phase of light is formed in a central part of said second region.

37. A mask for manufacturing a semiconductor device according to claim 36, wherein the depth d of said first groove for shifting phase of light is,

d=.lambda./(2(n.sub.1 -1))

.lambda.: wavelength of irradiation light

n.sub.1 : refractive index of said transparent substrate.

Description Submit all comments and votes

BACKGROUND OF THE INVENTION

The present invention relates to a mask for use in photolithography and its manufacturing technique, and particularly to a technique effectively applicable to a mask for use of manufacturing semiconductor integrated circuit device.

In recent years, very fine elements constituting a circuit, very fine wirings and very narrow spaces between the elements and wirings have been developed in semiconductor integrated circuit devices.

However, along with such development of the elements and wirings and of the spaces between elements and wirings, there arises a problem in that the accuracy of mask pattern transfer is lowered when an integrated circuit pattern is transferred onto a wafer by coherent light.

This problem will subsequently be described with reference to FIGS. 24(a)-(d).

When a given integrated circuit pattern formed on a mask 50 shown in FIG. 24(a) is transferred onto a wafer by a method of projection exposure or the like, the phases of lights each transmitted through each of a pair of transmission regions P.sub.1, P.sub.2 having light shield region N therebetween are identical to each other as shown in FIG. 24(b). Consequently, these interferential lights increase their intensities in light shield region N located between the above-mentioned pair of transmission regions P.sub.1, P.sub.2 as shown in FIG. 24(c). As a result, as shown in FIG. 24(d) the contrast of a projected image on a wafer is not only lowered, but also the depth of focus becomes shallow, causing the transfer accuracy of the mask pattern to be considerably lowered.

As a means to counteract these problems, a technique of phase shifting lithography has been developed, whereby the phase of light transmitted through the mask is controlled so as to improve the resolution and contrast of the projected image. The phase shifting lithography technique is disclosed, for example, in Japanese Laid-Open Patent No. 173744/1983 and Japanese Laid-Open Patent No. 67514/1987.

In the above-mentioned Japanese Laid-Open Patent No. 173744/1983 there is described the structure of a mask having a light shield region and a pair of transmission regions, wherein a transparent material is arranged at least in either one of the transmission regions sandwiching the light shield region therebetween, allowing a phase difference to be generated between the lights each transmitted through each of transmission regions at the time of exposure and thus these lights being interfered with each other to weaken themselves in the region on a wafer which should primarily be a light shield region.

The function of the light transmitted through such a mask as above will subsequently be described with reference to FIGS. 25(a)-(d).

When a given integrated circuit pattern formed on a mask 51 shown in FIG. 25(a) is transferred onto a wafer by the method of projection exposure or the like, a phase difference of 180.degree. is generated between the phase of light transmitted through a transmission region P.sub.2 having transparent material 52 of a pair of transmission regions P.sub.1, P.sub.2 which have light shield region N sandwiched therebetween and the phase of light transmitted through the normal transmission region P.sub.1 as shown in FIGS. 25(b) and (c). Therefore, the lights transmitted through the pair of transmission regions P.sub.1, P.sub.2 interfere with each other to offset them in light shield region N located between these transmission regions P.sub.1, P.sub.2. Consequently, as shown in FIG. 25(d), the contrast of a projected image on a wafer is improved. Thus, the resolution and depth of focus is improved, resulting in a higher accuracy of pattern transfer of the mask 51.

Also, in the above-mentioned Japanese Laid-Open Patent No. 67514/1987, there is described the structure of a mask having a light shield region formed by light shielding film and a transmission region formed by removing the light shielding film, wherein a fine aperture pattern is formed by removing a part of shielding film and at the same time, a phase shifting layer is provided on either one of the transmission region or the aperture pattern, and thus a phase difference is generated between the lights transmitted through the transmission region and the aperture pattern, preventing the distribution of amplitude of light transmitted through the transmission region from being spread in the horizontal direction.

SUMMARY OF THE INVENTION

Nevertheless, the present inventor has found that the conventional technique disclosed in the above-mentioned Japanese Laid-Open Patent No. 173744/1983 has the following problem:

The above-mentioned conventional technique in which a phase difference is generated between the lights transmitted through the pair of transmission regions does not has any problem as far as a pattern is simply and unidimensionally arranged in a repetitive manner. However, in the case that the pattern is complicated as in an actual integrated circuit pattern, the arrangement of the transparent material may be impossible, and a problem arises in that sufficient resolution is not obtained at some sections.

For example, in the case of an integrated circuit pattern 53 shown in FIG. 26. If transparent material is arranged in a transmission region P.sub.2, the resolutions in light shield regions N.sub.1 and N.sub.2 are certainly improved. However, if transparent material is arranged in transmission region P.sub.1 in order to improve the resolution in light shield region N.sub.3, the lights transmitted through transmission regions P.sub.1, P.sub.2 will have an identical phase, causing the resolution in light shield region N.sub.2 to be lowered. Also, in order to improve the resolution in the light shield region N.sub.3, a transparent material should be provided in such a transmission region as the transmission region P.sub.3. Then, the transparent material can be arranged in a part of transmission region P.sub.3. In such a case, however, there appears the reversing of phases in the lights transmitted through the same transmission region P.sub.3, and an unwanted pattern is formed on a wafer. Consequently, it becomes impossible to improve the resolution in light shield region N.sub.3.

Furthermore, if the pattern is complicated like an actual integrated circuit one, the arrangement of transparent material is restricted as mentioned above. This makes it difficult to prepare the pattern data of the transparent material. Conventionally, therefore, the pattern of the transparent material should be produced specially while taking into consideration the above-mentioned restriction on the arrangement when a mask having means for shifting phase of light is manufactured.

On the other hand, the known technique disclosed in Japanese Laid-Open Patent No. 67514/1987, whereby an aperture pattern is formed in a light shield region so as to generate a phase contrast between the light transmitted through the aperture pattern and the one transmitted through the transmission region, makes it difficult to arrange the aperture pattern, the same as in the case of the above-mentioned publication, if a pattern is as complicated and extremely fine as an actual integrated circuit pattern. For example, should the width of pattern of light shield region become narrower, there arises a problem in that the arrangement of an aperture pattern is difficult.

Furthermore, in this conventional technique, no consideration is given as to the lowering of light intensity at the corners of transmission region which takes place along with a further miniaturization of transmission region required, resulting in a problem posed in that the corners of a projected image are rounded.

The present invention is to solve the above-mentioned problems, and the object thereof is to provide a technique whereby the transfer accuracy of a pattern formed on a mask can be improved.

Another object of the present invention is to provide a technique whereby the manufacturing time of a mask having means for shifting phase of light can be reduced.

Still another object of the present invention is to provide a technique whereby the resolution of not only each side of a projected image but also of each corner thereof can be improved.

Among the inventions to be disclosed in the present application, those typical ones will subsequently be described.

Now, the first invention is a mask having light shield and transmission regions and transferring a given pattern at least by irradiation of coherent light locally, wherein a phase shifting portion is formed in a part of the aforementioned transmission region for shifting a phase of light transmitted, and a phase contrast is generated between the light transmitted through the aforementioned phase shifting portion and the light transmitted through the transmission region where the aforementioned phase shifting portion is not formed, and the aforementioned phase shifting portion is so arranged that the interferential lights of the aforementioned lights can weaken themselves in the boundary area of the aforementioned transmission and light shield regions.

The second invention is the method of manufacturing a mask wherein the pattern data of the phase shifting portion can automatically be prepared in accordance with the pattern data of the light shield region.

The third invention is a mask in which light shield and transmission regions are provided on a mask substrate and a given pattern in the mask is transferred at least by the irradiation of coherent light locally, wherein a groove having a depth to reach the main surface of the aforementioned mask substrate is formed, and simultaneously a phase contrast is generated between the light transmitted through the aforementioned groove and the light transmitted through the aforementioned transmission region, and a phase shifting groove is formed on the aforementioned mask substrate located below the aforementioned groove so as to allow the interferential lights of the aforementioned lights to weaken themselves at the end portion of the aforementioned light shield region.

The fourth invention is a mask in which light shield and transmission regions are provided on a mask substrate, and a given pattern is transferred at least by the irradiation of coherent light locally, wherein a groove having a depth to reach the main surface of the aforementioned mask substrate is formed in a part of the aforementioned light shield region, and a phase contrast is generated between the light transmitted through the aforementioned groove and the light transmitted through the aforementioned transmission region, and a transparent film is provided above the aforementioned groove so as to allow the interferential lights of the aforementioned lights to weaken themselves at the end portion of the aforementioned light shield region, and simultaneously, subtransmission regions are formed at the corners of the aforementioned transmission region.

According to the first invention mentioned above, the light transmitted through the phase shifting portion and the light transmitted through the portion where it is not formed interfere with each other to weaken themselves at the boundary portion of transmission and light shield regions so that the bleeding of contour of an image projected on a wafer can be reduced, and the contrast of the projected image can be improved considerably, resulting in a remarkable improvement of the resolution and depth of focus.

Particularly, in the present invention, no restriction on the arrangement of phase shifting portion takes place no matter how complicated the pattern is on the mask. Also, there is no difficulty in arranging the phase shifting portion no matter how narrow the width of pattern becomes in the light shield region.

According to the second invention mentioned above, the manufacturing time of a mask having means for shifting the phase of light can be reduced considerably because there is no need for preparing specially any pattern data of transparent film or phase shifting groove.

According to the third invention mentioned above, the light transmitted through the groove and phase shifting groove interfere with each other to weaken themselves, making it possible to reduce the bleeding of contour of a projected image and to improve the contrast thereof so that the solution and depth of focus can be improved remarkably.

According to the fourth invention mentioned above, the light intensity at the corner of a transmission region increases by the arrangement of a sub-transmission region thereat so that not only the resolution at each side of a projected image, but also the resolution at the corner thereof, can be improved.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a sectional view of the principal part of a mask embodying the present invention,

FIGS. 2(a)-(c) are sectional views illustrating the principal part of the mask in the respective processes of the manufacture thereof,

FIG. 3(a) is a sectional view of the mask shown in FIG. 1 in a state of exposure,

FIGS. 3(b)-(d) are diagrams representing the amplitude and intensity of the light being transmitted through the transmission region of the mask,

FIG. 4 is a sectional view of the principal part of another mask embodying the present invention,

FIG. 5(a) is a sectional view of the mask shown in FIG. 4 in a state of exposure,

FIGS. 5(b)-(d) are diagrams representing the amplitude and intensity of the light being transmitted through the transmission region of the mask shown in FIG. 4,

FIG. 6 is a sectional view of the principal part of another mask embodying the present invention,

FIG. 7 shows top plan views of the principal part of the mask shown in FIG. 6,

FIG. 8 illustrates the construction of a focused ion beam system employed in the manufacture of the mask,

FIGS. 9(a) and (b) are sectional views illustrating the principal part of the mask in the respective processes of the manufacture thereof,

FIG. 10 is a flow chart representing the procedures through which the pattern data for a phase shifting groove are prepared,

FIG. 11(a) is a sectional view of the mask shown in FIG. 6 in a state of exposure,

FIGS. 11(b)-(d) are diagrams representing the amplitude and intensity of the light being transmitted through the transmission region of the mask shown in FIG. 6,

FIG. 12 is a sectional view of the principal part of another mask embodying the present invention,

FIG. 13(a) is a sectional view of the mask shown in FIG. 12 in a state of exposure,

FIGS. 13(b)-(d) are diagrams representing the amplitude and intensity of the light being transmitted through the transmission region of the mask shown in FIG. 12,

FIG. 14 is a sectional view of the principal part of still another mask embodying the present invention;

FIG. 15 is a top plan view of the principal part of the mask shown in FIG. 14,

FIG. 16 is a top plan view of the principal part of a mask showing an example of pattern data for a groove and a sub-transmission region.

FIG. 17 is a flow chart representing the procedures through which the pattern data for the groove and sub-transmission region shown in FIG. 16 are prepared,

FIGS. 18(a)-(i) are illustrations showing the shapes of the pattern in the course of forming the pattern for the groove and sub-transmission region shown in FIG. 16,

FIG. 19(a) is a sectional view of the mask shown in FIGS. 14 and 15 in a state of exposure,

FIGS. 19(b)-(d) are diagrams representing the amplitude and intensity of the light being transmitted through the transmission region of the mask shown in FIGS. 14 and 15,

FIG. 20 is a sectional view of the principal part of still another mask embodying the present invention,

FIG. 21 is a top plan view of the principal part of the mask,

FIG. 22(a) is a sectional view of the mask shown in FIGS. 20 and 21 in a state of exposure,

FIGS. 22(b)-(d) are diagrams representing the amplitude and intensity of the light being transmitted through the transmission region,

FIG. 23 is a sectional view of the principal part of still another mask embodying the present invention,

FIG. 24(a) is a sectional view of a conventional mask in a state of exposure,

FIGS. 24(b)-(d) are diagrams representing the amplitude and intensity of the light being transmitted through the transmission region of the conventional mask,

FIG. 25(a) is a sectional view of a conventional mask in a state of exposure,

FIGS. 25(b)-(d) are diagrams representing the amplitude and intensity of the light being transmitted through the transmission region of the conventional mask,

FIG. 26 is a partial top plan view illustrating a part of the conventional mask.

DESCRIPTION OF A PREFERRED EMBODIMENT

Embodiment 1

FIG. 1 is a sectional view of the principal part of a mask embodying the present invention, and FIGS. 2(a)-(c) are sectional views illustrating the principal part of the mask in the respective processes of the manufacture thereof, FIG. 3(a) is a sectional view of the mask shown in FIG. 1 in a state of exposure, and FIGS. 3(b)-(d) are diagrams representing the amplitude and intensity of the light being transmitted through the transmission region of the mask.

Mask 1a shown in FIG. 1 of Embodiment 1 is, for example, a reticle used at a given process in manufacturing a semiconductor integrated circuit device. Also, the original pattern of an integrated circuit, which is, for example, five times its actual dimensions, is formed on the mask 1a of Embodiment 1.

Transparent mask substrate (hereinafter referred to simply as substrate) 2 constituting mask 1a is made of synthetic quartz glass or the like having, for example, a refractive index of 1.47. On the main surface of substrate 2, metal layer 3 of, for example, 500-3,000 .ANG. thick is patterned in a given shape.

Metal layer 3 comprises, for example, a Cr film so that it will constitute a light shield region A at the time of exposure. Also, metal layer 3 may be of a laminating construction, having chromium oxide laminated on the upper surface of a Cr layer.

Furthermore, the portion where the metal layer 3 has been removed constitutes a transmission region B, and the original pattern of the integrated circuit formed on mask 1a comprises of the light shield region A and the transmission region B.

In mask 1a of Embodiment 1, transparent film 4a formed in a pattern slightly wider than that of the above-mentioned pattern of metal layer 3 is provided on mask 1a in such a manner that its part extends over the contour of metal layer 3 into the transmission region B. Consequently, one transmission region B comprises of the region which is covered by transparent film 4a and the region which is not covered thereby.

Transparent film 4a is formed, for example, with indium oxide (InOx). A material having a sufficiently high transmittance against substrate 2 (at least more than 90% required) as well as an excellent property of adhesiveness to substrate 2 should be selected as a material for transparent film 4a. The width of the excessive portion of transparent film 4a is approximately 0.5 .mu.m, provided that, for example, the width of pattern of transmission region B is approximately 2.mu.m. Given the thickness of transparent film 4a measured from the main surface of substrate 2 is X.sub.1, the refractive index of transparent film 4a is n.sub.1, and the wavelength of light irradiated at the time of exposure is .lambda., transparent film 4a is formed to satisfy the relation of X.sub.1 =.lambda./[2(n.sub.1 -1)]. This relation is maintained in order to generate a phase contrast of 180.degree. at the time of exposure between the light transmitted through transparent film 4a and the one through the normal transmission region B in one transmission region B. For example, if the wavelength .lambda. of light irradiated at the time of exposure is 0.365 .mu.m (i line) and the refractive index n.sub.1 of transparent film 4a is 1.5, the thickness X.sub.1 of transparent film 4a measured from the main surface of substrate 2 will be approximately 0.37 .mu.m. Although not shown in the drawing, an alignment mark is provided on mask 1a to align its position with that of metal layer 3 when, for example, transparent film 4a is formed.

Next, the method of manufacturing mask 1a of Embodiment 1 will be described with reference to FIGS. 2(a)-(c).

First, the surface of transparent substrate 2 made of synthetic quartz glass or the like is ground, and after it is cleaned, metal layer 3 made of Cr or the like of, for example, approximately 500-3,000 .ANG. is formed on its main surface by sputtering or others as shown in FIG. 2(a). Subsequently, on the metal layer 3, photoresist (hereinafter referred to as resist) 5a of, for example, 0.4-0.8 .mu.m will be coated. Then, after resist 5a is prebaked, a given portion of 5a is irradiated with electron beam E by the electron beam exposure method or the like in accordance with the integrated circuit pattern data with the integrated circuit pattern data of a semiconductor integrated circuit device coded and stored in advance in a magnetic tape which is not shown in the drawing. In the integrated circuit pattern data, the position coordinate, shape or others are also stored.

Next, as shown in FIG. 2(b), the exposed portion of resist 5a, for example, is removed by a developing solution. Then, the exposed metal layer 3 is removed by etching with a dry etching method or the like to gain the pattern of a given shape.

Subsequently, resist 5a is removed by an exfoliative agent and substrate 2 is cleaned. Then, after inspection, as shown in FIG. 2(c), the main surface of substrate 2 is covered with transparent film 4a of indium oxide (InOx) or others by a sputtering method or the like. At this juncture, the thickness X.sub.1 of the transparent film measured from the main surface of substrate 2 is, for example, approximately 0.37 .mu.m.

After that, the upper surface of transparent film 4a is coated with resist 5b of, for example, 0.4-0.8 .mu.m and furthermore, on its upper surface, an aluminum antistatic layer 6 of, for example, 0.05 .mu.m thick is formed by a sputtering method or the like.

Subsequently, the pattern formed on transparent film 4a will be transferred to resist 5b by an electron beam exposure system or the like in accordance with the pattern data of transparent film 4a.

The pattern data of transparent film 4a is automatically formed by magnifying or demagnifying the light shield region A or transmission region B of the above-mentioned pattern data of the integrated circuit. In embodiment 1, for example, the pattern data of transparent film 4a is automatically formed by thickening the pattern width of light shield region A by, for example, approximately 0.5-2.0 .mu.m.

Then, mask 1a shown in FIG. 1 will be produced through the processes of development, etching of a predetermined portion of transparent film 4a, removal of resist 5b, cleaning, inspection and others.

In order to transfer the pattern of the integrated circuit formed on mask 1a onto a wafer covered with resist by use of the mask 1a thus produced, the following steps will be taken, for example.

Mask 1a and a wafer are set on a projection aligner for demagnification (which is not shown in the drawing), whereby the original pattern of integrated circuit on mask 1a is projected on a wafer after it has been reduced to 1/5 optically, and each time the wafer is sequentially moved step by step, projection and exposure are repeated so that the pattern of the integrated circuit formed on mask 1a can be transferred onto the entire surface of the wafer.

Next, the function of Embodiment 1 will be described with reference to FIGS. 3(a)-(d).

In mask 1a of Embodiment 1 shown in FIG. 3(a), a phase contrast of 180.degree. is generated between the light transmitted through transparent film 4a and the one through the normal transmission region B (FIGS. 3(b) and (c)) in each of transmission regions B of mask 1a when the original pattern of a given integrated circuit formed on mask 1a is transferred onto a wafer by a projection aligner for demagnification or others.

Here, the lights transmitted through the same transmission region B having reverse phases respectively weaken each other at the boundary area of transmission region A and light shield region B because transparent film 4a is arranged around transmission region B. As a result, the bleeding of a contour of image projected on the wafer is reduced so as to improve the contrast of the projected image considerably. Thus, the resolution and depth of focus will remarkably be improved (FIG. 3(d)). Also, since light intensity is the square of light amplitude, the waveform in a negative side is reversed to a positive side as shown in FIG. 3(d).

In this way, it is possible to attain the following effect according to Embodiment 1:

(1). While lights are being transmitted through each of the transmission regions B at the time of exposure, a phase contrast of 180.degree. is generated between the light transmitted through transparent film 4a and the one through the region which is not covered with transparent film 4a. These lights thus transmitted weaken each other at the boundary area of light shield region A and transmission region B, making it possible to reduce the bleeding of contour of image projected on the wafer. As a result, the contrast of the projected image is considerably improved so that the resolution and depth of focus can be improved remarkably.

(2). Through (1) mentioned above, more tolerance for exposure can be allowed.

(3) Since a phase contrast can be generated within one transmission region B, the arrangement of transparent film 4a is not restricted no matter how complicated the pattern is on mask 1a. Also, the arrangement of transparent film 4a can be made with ease no matter how narrow the pattern width is in light shield region A. Consequently, the pattern transfer accuracy will never be lowered locally even when a pattern formed on mask 1a is highly complicated and fine like the pattern of an integrated circuit so that the pattern transfer accuracy of the entire pattern formed on mask 1a can be improved remarkably.

(4). The pattern data of transparent film 4a can be obtained automatically based on the pattern data of light shield region A or transmission region B, making it possible to prepare the pattern data of transparent film 4a in a short period of time and also with ease. As a result, a remarkable reduction in time required for the manufacture of a phase shifting mask can be achieved.

Embodiment 2

FIG. 4 is a sectional view of the principal part of another mask embodying the present invention, FIG. 5(a) is a sectional view of the mask shown in FIG. 4 in a state of exposure, and FIGS. 5(b)-(d) are diagrams representing the amplitude and intensity of the light being transmitted through the transmission region of the mask in FIG. 4.

In mask 1b of Embodiment 2 shown in FIG. 4, transparent film 4b is arranged in the vicinity of the central part of transmission region B.

In this case, too, transparent film 4b is formed on substrate 2 in such a manner that its thickness X.sub.1 will satisfy the relation of X.sub.1 =.lambda./[2(n.sub.1 -1)] so as to generate as shown in FIGS. 5(b)-(d) the phase contrast of 180.degree. between the light transmitted through transparent film 4b and the one through the normal transmission region B in each of transmission regions B, B of mask 1b (FIGS. 5(b) and (c)). These lights thus transmitted will weaken each other at the boundary area of transmission region B and adjacent light shield region A, A so that the bleeding of contour of the image projected on a wafer can be reduced. As a result, the contrast of a projected image can be considerably improved, making it possible to improve the resolution and depth of focus remarkably (FIG. 5(d)).

Also, in this case, the pattern data of transparent film 4b can automatically be prepared by thinning by a given dimension the width of a pattern of transmission region B which has been obtained by reversing the positive pattern data of, for example, an integrated circuit pattern to the negative one.

According to Embodiment 2, therefore, the same effects as in the aforementioned Embodiment 1 can be attained.

Embodiment 3

FIG. 6 is a sectional view of the principal part of another mask embodying the present invention, FIG. 7 shows top plan views of the principal part of the mask shown in FIG. 6, FIG. 8 illustrates the construction of a focused ion beam system, FIGS. 9(a) and (b) are sectional views illustrating the principal part of the mask in the respective processes of the manufacture thereof. FIG. 10 is a flow chart representing the procedures through which the pattern data for a phase shifting groove are obtainable, FIG. 11(a) is a section view of the mask shown in FIG. 6 in a state of exposure, and FIGS. 11(b)-(d) are diagrams representing the amplitude and intensity of the light being transmitted through the transmission region of the mask shown in FIG. 6.

The mask of Embodiment 3 will subsequently be described with reference to FIGS. 6 and 7. The crosshatching in FIG. 7 shows the light shield region A.

In mask 1c of Embodiment 3, phase shifting groove 7a is formed on substrate 2 instead of transparent film 4a of aforementioned Embodiment 1 as a means for generating a phase contrast between the lights being transmitted through transmission region B at the time of exposure.

Phase shifting groove 7a is arranged around transmission region B. In other words, phase shifting groove 7a is arranged along the contour of metal layer 3. The width of phase shifting groove 7a is approximately 0.5 .mu.m if the pattern width of transmission region B is given to be, for example, approximately 2.0 .mu.m. Then, phase shifting groove 7a is formed to satisfy the relation of d=.lambda./[2(n.sub.2 -1)] where d is the depth thereof, n.sub.2 is the refractive index of substrate 2, and .lambda. is the wavelength of light irradiated at the time of exposure. This relation is maintained to generate a phase contrast of 180.degree. between the phase of light transmitted through phase shifting groove 7a and that of light through the normal transmission region B in the lights being transmitted through each of transmission regions B at the time of exposure. For example, in case where the wavelength .lambda. of the light irradiated at the time of exposure is given to be 0.365 .mu.m (i line), the depth d of phase shifting groove 7a can be approximately 0.39 .mu.m. Also, although not shown in the drawing, an alignment mark is provided on mask 1c when, for example, phase shifting groove 7a is formed to align its position with that of metal layer 3.

Next, focused ion beam system 8 for use in manufacturing mask 1c will be described with reference to FIG. 8.

Inside ion source 9 installed above the system body, such a dissolved liquid metal as gallium (Ga) or the like, for example, is contained, though not shown in the drawing. Below ion source 9, withdrawable electrode 10 is installed, beneath which is provided the first lens electrode 11a and the first aperture electrode 12a formed by static lens. Below aperture electrode 12a, the second lens electrode 11b, the second aperture electrode 12b, blanking electrodes 13 for controlling the on/off of beam irradiation, and then third aperture electrode 12c and deflection electrode 14 are installed.

With the formation of each electrode thus installed, the ion beam emitted from ion source 9 is irradiated under the controls of the above-mentioned blanking electrodes 13 and deflection electrode 14 onto the mask 1c which is held by holder 15 with pattern yet to be formed thereon. Then, metal layer 3 or substrate 2 can be processed by etching with the ion beam by setting at the time of scanning beforehand its irradiation time and scanning numbers per unit of pixel of, for example, 0.02.times.0.02 .mu.m.

Holder 15 is installed on the sample stand 16 movable in the directions of X and Y, and sample stand 16 can be positioned by sample stand driving motor 19 when its position is recognized by laser interferometer 18 through laser mirror 17 installed at the side of the sample stand. Also, above holder 15, there is installed secondary ion and secondary electron detector 20 so that the secondary ion and secondary electron generated by a workpiece can be detected. Furthermore, electron shawer radiating member 21 is installed above secondary ion and secondary electron detector 20 to prevent a workpiece from being electrified.

The inside of the processing system set forth above has a structure keeping itself under vacuum by vacuum pump 22 shown in the drawing below the above-mentioned sample stand 16. Also, each of the processing systems mentioned above is structurally controlled by each of the control members 23-27 installed outside the system body so that their operations are controlled, and each of the control members 23-27 is also controlled by controlling computer 33 through each of the interface members 28-32. Controlling computer 33 has terminal 34, magnetic disk unit 35 to record data, and MT deck 36.

Next, the method of manufacturing mask 1c of Embodiment 3 will be described with reference to FIG. 8, FIGS. 9(a) and (b) and FIG. 10.

First, as shown in FIG. 9(a), metal layer 3 of, for example, 500-3,000 .ANG. is formed by sputtering or the like on the main surface of substrate 2 which has been ground and cleaned. Then, mask 1c is held by holder 15 of focused ion beam system 8.

Next, an ion beam is charged from ion source 9. This ion beam is converged by each of the above-mentioned electrodes into a beam diameter of, for example, 0.5 .mu.m. Then, an ion beam current of approximately 1.5 .mu.A is obtained. Subsequently, this focused ion beam is irradiated onto a given portion of metal layer 3 in accordance with the pattern data of an integrated circuit pattern stored beforehand in a magnetic tape of MT deck 36. Then, metal layer 3 is etched. At this juncture, the irradiation time per pixel is, for example, 3.times.10.sup.-6 second, and the scanning numbers are approximately 30. Thus, as shown in FIG. 9(b), metal layer 3 is patterned. The patterning of metal layer 3 may also be carried out by an electron beam exposure method or the like as in the aforementioned Embodiment 1.

After this, a given quantity of ion beam is irradiated onto the alignment mark provided on mask 1c, which is not shown in the drawing, so as to detect a generated secondary electron by secondary ion secondary electron detector 20, and the position coordinates are computed in accordance with the detection data.

Then, based on the position coordinates of the alignment mark thus worked out, sample stand 16 is moved so that the ion beam can be irradiated onto the location where phase shifting groove 7a is formed.

Next, in accordance with the pattern data of phase shifting groove 7a, the ion beam is irradiated onto substrate 2, which has been exposed by the pattern formation of metal layer 3, along the contour of metal layer to form phase shifting groove 7a (FIG. 6). At this juncture, the depth, width and others of phase shifting groove 7a can be defined by the focused ion beam accurately with ease.

The pattern data of phase shifting groove 7a is prepared by a logic arithmetic operation on the pattern data of light shield region A (or transmission region B) and the pattern data obtainable by magnifying or demagnifying the pattern data of light shield region A (or transmission region B).

For example, as shown in FIG. 10, the pattern data of an integrated circuit is first prepared through the process of LSI circuit design (step 101a), CAD design data (101b), and Boolean OR (101c), and then the data is produced by the sizing process (102) for the patterned data having the pattern width of light shield region A which has been thickened only by a given dimension. At the same time, the data for pattern data of transmission region B are prepared in the process of reverse tone (103) by reversing the positive pattern data of an integrated circuit pattern into the negative one. Then, the pattern data of phase shifting groove 7a are automatically prepared (105) by executing AND of these pattern data (104).

Next, after phase shifting groove 7a has been formed, the bottom face of phase shifting groove 7a formed on mask 1c i