Data processing: generic control systems or specific application – Specific application – apparatus or process – Product assembly or manufacturing
Reexamination Certificate
1998-04-08
2002-04-09
Wiley, David (Department: 2155)
Data processing: generic control systems or specific application
Specific application, apparatus or process
Product assembly or manufacturing
C716S030000, C716S030000, C709S245000, C709S227000, C709S241000, C709S241000
Reexamination Certificate
active
06370441
ABSTRACT:
BACKGROUND OF THE INVENTION AND RELATED ART
The present invention relates to a method of correcting designed-pattern data obtained by data-processing a plurality of designed patterns, a method of electron beam exposure including the method of correcting designed-pattern data, a photomask, a method of fabrication of a photomask, a method of optical exposure, a semiconductor device, a method of fabrication of a semiconductor device, and an apparatus for correcting designed-pattern data.
A photomask used in the process of optical lithography for the fabrication of a semiconductor device has a structure in which a patterned light-shielding thin layer or light semi-shielding thin layer is formed on a substrate transparent to exposure light, such as a glass substrate. In the fabrication of the semiconductor device, it is required to transfer a pattern formed in the photomask to a photo resist formed, e.g., on a semiconductor substrate. For this purpose, it is required to produce data from a plurality of designed patterns designed by means of CAD or the like and expose an electron beam resist formed on a mask blank with an electron beam (electron beam exposure) on the basis of pattern data for electron beam exposure prepared from the produced data so that patterns corresponding to the designed patterns are faithfully formed in the electron beam resist. In some case hereinafter, patterns formed in a photomask will be referred to as “mask patterns”, pattern data obtained by data-processing a plurality of the designed patterns will be referred to as “designed-pattern data”, pattern data for electron beam exposure will be referred to as “pattern data for exposure”, patterns formed in an electron beam resist will be referred to as “resist patterns”, and patterns formed in a photo resist will be referred to as “printed patterns”. For example, the designed-pattern data for fabricating a photomask is constituted of a steam format called GDSII/Stream in which the designed patterns are represented by polygons or of an electron beam data format (“EB data format” hereinafter) in which the designed patterns are represented by rectangles and trapezoids.
Meanwhile, in electron beam exposure, a phenomenon called “proximity effect” controls the limitation of resolution for forming mask patterns, and it is a big problem when a photomask requiring micro- or nano-processing is fabricated. The above proximity effect takes place by scattering of electrons in a solid, and it can be classified into two categories depending upon the shapes and/or geometrical layouts of the mask patterns to be formed. That is, it is classified into an intra proximity effect which takes place in an isolated small mask pattern and an inter proximity effect which takes place between adjacent mask patterns. In the intra proximity effect, electrons incident on an electron beam resist are scattered into portions of the electron beam resist other than portions of the electron beam resist where resist patterns are to be formed. As a result, accumulated energy in the portion of the electron beam resist where resist patterns are to be formed cannot reach a predetermined threshold value, so that the sizes of the resist patterns are smaller than those of their designed patterns, or that corner portions of the resist patterns are rounded to some extent. On the other hand, in the inter proximity effect, accumulated energy in resist patterns to be formed in an electron beam resist reaches a predetermined threshold value due to the scattering of electrons from resist pattern(s) close or adjacent thereto, and the resist patterns have larger dimensions than their designed patterns or, in the worst case, resist patterns are in contact with another. In the direct exposure of an electron beam resist formed on a semiconductor substrate with an electron beam, the proximity effect similarly causes the above problems.
For correcting the proximity effect with regard to an electron beam exposure apparatus according to a variable shaped beam method, there is proposed a method in which electron beam dosages (“dosage”, hereinafter) are varied depending upon each resist pattern to be formed in an electron beam resist. In the above method, evaluation points (indicated by black dots) are provided on edges of a designed pattern as schematically shown in
FIG. 10A
, accumulated energy determined on the basis of the EID (Exposure Intensity Distribution) function proposed by T. H. P. Chang, J. Vac. Soi. Technol. 12, 1271 (1983) is calculated with regard to each evaluation portion, and an optimum dosage for forming a resist pattern is determined. However, when geometrical layouts of adjacent designed patterns in the vicinity of the evaluation points differ, one obtained optimum dosage in one evaluation point differs from another obtained optimum dosage in another evaluation point, and eventually, a weighted mean dosage is inevitably determined to be an optimum dosage. Accuracy is therefore different from one evaluation point to another, or no desired accuracy is secured in some cases. For overcoming the above defect, it is required to divide a designed pattern as shown in FIG.
10
B and control the dosage with regard to each evaluation point. Numerals in
FIGS. 10A and 10B
show the dosages. However, when all of designed patterns are divided to smaller sizes, the time required for calculation of the proximity effect correction increases, and the file size (data volume) of the designed-pattern data or the time required for electron beam exposure increases. The consequent problem is that the fabrication of a photomask takes a long time.
On the other hand, even if the proximity effect correction is carried out to form the mask patterns in order to accurately transfer the designed patterns to a photo resist, there is another problem that an optical proximity effect takes place when the mask patterns are transferred to a photo resist formed, e.g., on a semiconductor substrate by means of exposure light through the photomask, so that the shapes of the patterns formed in the photo resist (printed patterns) differ from those of their designed patterns. That is, the optical exposure process in the production process of a semiconductor device has the problem of an optical proximity effect. That is, when a mask pattern whose size approximates a wavelength of the exposure light is transferred to a photo resist, an interference phenomenon with the exposure light remarkably takes place, and a printed pattern differs from its designed pattern in size. The optical proximity effect results in phenomena in which a line width of an isolated line differs from a line width of a repetitive line and in which line-end shortening occurs, and it causes the deterioration of gate line width controllability or a decrease in an alignment margin. As a result, transistors vary in characteristics, and eventually, the yield of chip production decreases. The optical proximity effect has extremely detrimental effects on the efficiency of production of semiconductor devices. Since the above problem is fatal particularly to memory cells which have many repetitive patterns and are required to have high-degree integration, there has been developed an advanced automatic optical proximity effect correction (OPC) system on the basis of a light intensity simulation for memory cells which come after the generation of design rule 0.35 &mgr;m.
The optical proximity effect is also classified into an intra optical proximity effect and an inter optical proximity effect. In the intra optical proximity effect, the exposure light diffracts in a mask pattern itself, and as a result, the dimensions of a pattern formed in a photo resist by focusing the exposure light on the photo resist differ from those of its designed pattern, or, for example, the short side and the long side of a rectangle differ to a great extent in formation accuracy. On the other hand, in the inter optical proximity effect, exposure light which diffracts from a mask pattern adjacent to some other mask pattern interferes with exposure light whic
Backer Firmin
Kananen Ronald P.
Rader Fishman & Grauer
Sony Corporation
Wiley David
LandOfFree
Method and apparatus of correcting design-patterned data,... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus of correcting design-patterned data,..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus of correcting design-patterned data,... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2857854