JP3054611B2 - Electron beam writing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an electron beam writing method.

[0002]

2. Description of the Related Art In recent years, an electron beam drawing method of drawing a pattern by using an electron beam can draw a much finer pattern than a case of drawing a pattern by using light. It is about to become the mainstream of pattern drawing.

FIG. 6 shows the configuration of an electron beam writing apparatus used in this electron beam writing method. In FIG. 6, an electron beam emitted from an electron gun 52 is focused by a focusing lens 54 and is applied to a shaping aperture 56. The image of the shaping aperture 56 is formed on the shaping aperture 62 by the projection lens 58. Here, by adjusting the position of the image of the shaping aperture 56 on the shaping aperture 62 by the shaping deflector 60, a rectangular or triangular beam of an arbitrary size is obtained. The image of the shaping aperture 62 is reduced by the objective lens 66 and formed on the sample 80. Here, the beam is largely deflected by the high-precision main deflector 68, and the beam irradiation position is finely adjusted by the high-speed sub-deflector 64.

[0004]

In the electron beam writing method for writing a pattern using such an electron beam writing apparatus, there is a problem in that the pattern dimensional accuracy is deteriorated due to the long-distance photosensitive action described below. This will be described with reference to FIG.

Now, assume that an electron beam 90 is incident on the surface of a sample 80 as shown in FIG. Some of the electrons incident on the surface of the sample 80 are reflected, while other electrons are reflected on the sample 8.
Secondary electrons are generated from the zero surface. The reflected electrons and secondary electrons are transmitted to the objective lens 6 of the electron beam writing apparatus.
6, the secondary electrons 92 are emitted from the lower surface 66a while being reflected again. These re-reflected electrons or secondary electrons give energy to the resist 81 applied to the surface of the sample 80, and the resist 81 is also exposed to areas other than the point where the electron beam 90 is incident. This phenomenon is called a long-distance photosensitive action.

[0006] The range over which the long-distance photosensitivity is exerted extends over an area several tens of mm from the beam irradiation position. For this reason, when the average irradiation amount is largely distributed over a region of several mm or more on the surface of the sample 80, a large size distribution is brought about in the pattern size of the developed resist. This problem is particularly serious in a resist mask used for photolithography and X-ray lithography.

GHOST is used to correct this long-distance photosensitive action.
It is conceivable to apply the law. However, this GHOS
Since the T method is a proximity effect correction method, when the GHOST method is used, it is necessary to blur the beam from a few mm to several tens mm, which is about the extent of the long-distance photosensitive action, and this is not practical. Occurs.

It is also conceivable to perform correction by changing the irradiation amount for each shot of the electron beam. However, in this case, there is a problem that the irradiation amount data for each shot is required, and the load on the data processing computer is large, which is not practical.

As described above, the conventional electron beam writing method has a problem that a pattern dimension with high accuracy cannot be obtained due to a long-distance photosensitive action. There is a problem that it is not practical to use other methods to avoid this.

The present invention has been made in view of the above circumstances, and has as its object to provide an electron beam drawing method capable of obtaining a pattern size with high accuracy.

[0011]

According to the present invention, there is provided an electron beam lithography method according to the present invention, wherein a point Y in a writing area of a sample is irradiated with a unit dose of an electron beam at a point X in the writing area. Obtaining a dose density distribution ρ (X, Y) per unit dose by the action; and n (≧≧)
2) dividing into small areas i (1 ≦ i ≦ n), obtaining a dose D _i in each small area i based on a pattern to be drawn;
A step of obtaining a dose density F _i due to long-distance sensitization in each small area i based on (X, Y) and the dose D _i , and a step of obtaining a dose density F _i in each small area i of the region, determining a corrected dose density C _i to remove the influence of long-distance photosensitive action, the correction dose density C _i obtained for each small area i, as fill the corresponding small area The method includes a step of irradiating with an electron beam and a step of pattern drawing based on drawing data.

[0012] Incidentally, the step of obtaining the corrected dose density C _i in each small area i is corrected dose B _i for removing the influence of long-distance photosensitive action in each small region i
_Is determined so that F _i + B _i = 0, and an offset irradiation amount B _os is introduced, and B _i +
The offset dose B _{os is} set such that the value of B _os is non-negative.
And calculating the corrected dose density C _i as C _i = B _i
+ B _os may be provided.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of an electron beam writing method according to the present invention will be described with reference to the drawings. The electron beam writing method according to this embodiment is a writing method for removing the influence of a long-distance photosensitive action, and the processing procedure is shown in FIG.

First, before drawing by using an electron beam on a resist-coated sample, a long-distance photosensitive action
The dose density distribution ρ (X, Y) per unit dose is determined (see step F1 in FIG. 1). This dose density distribution ρ
(X, Y) is measured as follows. One point in the drawing area of the sample on which the resist has been applied is irradiated with an electron beam having a constant irradiation amount. Thereafter, development is performed, and the resist film thickness at a position other than the irradiation point is measured to determine the irradiation amount distribution due to the long-distance photosensitive action. Next, using another sample coated with a resist, an area having a certain area is irradiated with an electron beam having a certain irradiation amount smaller than the above-mentioned certain irradiation amount, and an irradiation amount distribution is obtained in the same manner as described above. . From the above two measurement results, the irradiation density distribution ρ per unit area due to the long-distance photosensitive action at the point X in the drawing area when one point Y in the drawing area is irradiated with the electron beam of the unit irradiation amount
(X, Y) is determined. The irradiation density distribution other than the irradiation point is obtained from the first measurement, and the irradiation density distribution at the irradiation point is obtained from the subsequent measurement.

The irradiation density distribution ρ (X, Y)
It is possible to obtain an approximation that can be considered to be determined only by the positional deviation (vector amount) XY between the measurement point X and the measurement point X. Hereinafter, this approximation will be used and expressed as ρ (XY). In many cases, ρ (XY) is the vector X−
The approximation may be made based on the absolute value of Y. The dose density distribution may be approximated by a Gaussian distribution, a multi-Gaussian distribution, a Lorentz distribution, a multi-Lorentz distribution, or a polynomial distribution. The dose density distribution ρ (X
−Y) may be determined at a plurality of different representative positions Y, and determined at other positions by interpolation.

Next, the drawing region of the sample on which the resist is to be drawn on which the resist is applied is divided into n (≧ 2) small regions (see step F2 in FIG. 1). The size of the divided small region is larger than the spread of the backscattering of electrons, and may be, for example, a fraction of the typical size (about 10 mm) of the dimensional variation due to the long-distance photosensitive action. For example, as shown in FIG.
The drawing area 20 of the sample 10 having a size of 130 mm × 130 mm, which is m, is divided into small square areas 25 having a size of 1 mm × 1 mm.

When the size of the small area is larger than the size of the deflection area of the electron beam writing apparatus used for pattern writing, the size of the small area is set to an integral multiple of the size of the deflection area. This is preferable for increasing the drawing efficiency. For example, if the deflection area is a square having a side of 500 μm, it is preferable to set the small area 25 to a square of 1 mm (= 1000 μm). As shown in FIG. 2A, the drawing area 20 is a square of 130 mm square and the small area 25
Is a square of 1 mm square, the number n of the small areas 25 is 1
6900 (= 130 × 130). In the small area 25 divided in this way, for example, numbers 1, 2,
(See FIG. 2B). Note that FIG.
FIG. 3A is an enlarged view in which the lower left portion of the divided drawing area 20 is enlarged, and each small area 25 is numbered.

[0018] Then obtain the dose D _i of the electron beams in each small area i (1 ≦ i ≦ n) on the basis of the drawing pattern to be drawn in the drawing area of the sample. Here, i indicates the number assigned to the small area. The dose D _i is a dose corresponding to the drawing pattern in the small area i.
As a method of obtaining the exposure dose D _i, expand the entire drawing data about the drawing pattern, but it is also possible to determine by calculating the sum of the areas of each shot contained in the small area i, the drawing data In the case of having a hierarchical structure, it is preferable from the viewpoint of calculation efficiency that areas included in a higher hierarchical structure are collectively handled as much as possible. Small area i
When the area of the pattern included in the S _i, the base dose density per unit area of the pattern drawing and Q _o, D _i =
Q _o · S _i .

Next, in each small area i (1 ≦ i ≦ n),
Exposure density F _i per unit area due to long-range photosensitivity
Ask for. This dose density F _i is obtained using the following equation.

F _i = Σρ (X _i −X _j ) · D _j where X _i and X _j indicate the positions of the representative points (for example, the center points) of the small areas i and j, and Σ indicates the index j. 1 to n
Up to the summation. According to the knowledge of the inventor, when a 50 KeV electron beam drawing apparatus is used, the irradiation density F _i is at most several% of the irradiation density when pattern writing is performed based on the writing data.

[0021] Then, based on the dose density F _i at each of the small areas i, the corresponding region i, obtaining a correction dose density C _i to remove the influence of long-distance photosensitive action (Fig. 1
Step F5).

[0022] The dose density C _i is determined as follows, for example. First, in each small area i, F _i + B _i =
A correction amount B _i (= −F _i ) that becomes 0 is obtained. Since the sign of F _i is positive, the sign of the correction amount B _i is negative. Since negative dose can not be defined, by introducing the offset value B _os, determining the offset value B _os such that the sum B _os + B _i and the offset value B _os and the correction amount B _i is not negative. There are various _ways to determine the offset value B _os , but the minimum value of B _i + B _os is made zero, that is, n values −B ₁ ,.
Maximum value M _ax of -B _n (-B _i) the offset value B
It is practical to use _os . Also, F _i = Σρ (X _{i
-X j) · (D j +} B j) as to obtain a correction amount B _i in a self-consistent manner is also possible.

The sum B _i + B _os of the offset value B _os and the correction amount B _i obtained in this manner is used as the dose density C _i for eliminating the influence of the long-distance photosensitive action in the small area i.
And

[0024] Next, each small area i (1 ≦ i ≦ n), and irradiated with an electron beam correction dose density C _i obtained is corrected drawing (see step F6 in Figure 1). This correction drawing is performed as follows. In this correction drawing, the correction dose density C _i in the small area i is constant within a small region i. For this reason, when writing is performed using a variable shaped beam type electron beam writing apparatus, the beam shape is set to the largest rectangle among rectangles each having a size that is a fraction of the dimension of the small area i. The correction drawing same in the correction dose density C _i in the small region 25 within 3, shots of the same dimension 2
At step 8, a solid drawing is performed. For example, the small area 25 is a square having a side of 1 mm, and the size of the shot 28 is 2.5 μm.
Then, the small area 25 can be painted out with 160000 (400 × 400) shots. At this time, if the discontinuity of the connection portion between shots of the electron beam cannot be ignored as a change in the resist dimension, it is effective to blur the electron beam to be irradiated (to widen the foot of the electron beam). In this case, it is preferable that the beam be blurred in such a manner that the blur is large. However, in order to largely blur the beam, it is necessary to largely shift the setting of the objective lens of the electron beam writing apparatus. For this reason, it is necessary to adjust the electron beam lithography apparatus, which causes a decrease in the operation efficiency of the electron beam lithography apparatus. Therefore, in practice, as shown in FIG. 4, the irradiation density of one shot is reduced to 1 / shot size.
If Bokase to about 2 (see graph g _1), the one shot position, overlap irradiation shots surrounding the shot position. Thus Synthetic dose density, can substantially constant what is obtained (see the graph g ₂ in FIG. 4), to obtain a continuous dose of the correction drawing. According to the knowledge of the present inventor, even if it is blurred by about several tens of μm, a practical allowable pattern size can be obtained. The distance u between the adjacent shot (the distance between the center of the shot and the center of the shot adjacent to the shot) u is the shot size S.
By making it smaller than, for example, S = 2u,
Making the adjacent shots overlap is also effective in obtaining continuity of the dose. Alternatively, the correction drawing may be performed a plurality of times, and the boundaries of shots may be shifted in different drawing times.

After the correction drawing is completed in this way,
Pattern drawing is performed on the sample on which the correction drawing has been performed based on the drawing data. Note that this pattern drawing may be performed before the corrected drawing, not after the corrected drawing.

As described above, according to the electron beam lithography method of the present embodiment, the irradiation amount F _i and the correction irradiation amount B _i due to the influence of the long-distance photosensitive action caused by pattern drawing.
Since the correction drawing is performed so that the sum of the constants is constant in all the small areas, the influence of the long-distance photosensitive action can be eliminated. For this reason, an accurate pattern dimension can be obtained. The pattern drawing method is practical.

In the above embodiment, the irradiation density F _i
May be calculated as follows. According to the inventor's knowledge, a typical length of a change in the dose density distribution ρ (X) (for example, a variance value when ρ (X) is a Gaussian distribution) is:
It is about the distance between the lower surface of the objective lens and the sample surface (the distance h shown in FIG. 7). Therefore, the length a of one side of the small area i
Must be smaller than the distance h.

If necessary, the size of the small area can be reduced. However, if the length a is set to about 0.1 _h , the variation of the irradiation density F _i due to the long-distance photosensitive action is several%. The error associated with the size of the small area is 1% or less, and the required accuracy is usually obtained. Further, the length a is set to 0.0
If it is set to 1 h, the dimensional fluctuation can be suppressed to 1 nm or less in principle. Even in this case, h = 10 mm, and the size of the small region can be 0.1 mm.

For this reason, as shown in FIG. 5 (a), when calculating the irradiation density F _i , two types of small regions, for example, 0.1 h and 0.01 h are prepared. The irradiation dose associated with pattern drawing is obtained for the small areas of each size. When calculating the dose density F _i , when calculating the effect of a small area that is more than 3 h away from the target small area i, for example, a coarsely divided small area (for example, a small area having a size of 0.1 h). Area), and when obtaining the effect of a small area within a distance of 3 h, a small area (for example, a small area having a size of 0.01 h) is used.
Calculation time can be saved. For example, as shown in FIG. 5 (b), if the small region i for which to calculate the F _i is in the center of the region 30, region 30 and region 32
Is used to calculate the effect of the small area inside the small area (a small area having a size of, for example, 0.01 h), and to calculate the influence of the other area 34, the area is roughly divided. Small area (small area of, for example, 0.1 h)
Is used, the calculation time of the dose density F _i can be saved.

Further, the irradiation density by the correction drawing can be determined as follows. Assuming that the dose density at the time of pattern writing is constant, the final dose density differs between when the pattern density on the sample is low and when the pattern density is high. Therefore, it is necessary to change the resist development conditions according to the irradiation amount. Therefore, F _i and C _i are obtained by using Q _o as the reference dose density for pattern drawing. Next, f
= Seek _{(Q o + F i + C} i) / Q o. Then, instead of Q _o, a reference dose density of Q _o / f again small area i. In this case, the dose density for pattern writing is a smaller value than the dose density when there is no long-distance photosensitive action. In the actual pattern writing, the irradiation density of the pattern writing is determined so that the proximity effect including the irradiation density by the correction writing can be corrected in each small area. Further, in calculating the dose density for proximity effect correction in the dose correction at the time of pattern drawing, the dose is calculated including the dose density at the time of pattern writing and the dose density at the time of pattern drawing.

In an electron beam writing apparatus, the electron beam irradiation amount cannot usually take a continuous value, but can take only a plurality of discrete values. Therefore, in the correction drawing of the small area i (1 ≦ i ≦ n),
The discrete value closest to the dose (= C _i × the area of the small area i) corresponding to the corrected dose density C _i obtained in step F6 of FIG. 1 is selected, and the selected discrete value is selected. The small area i is irradiated with the irradiation amount. Even in this case, the dimensional accuracy of the drawn pattern is not degraded.

This will be described below. According to the knowledge of the present inventor, when an electron beam lithography apparatus of 50 KeV is used, the irradiation density F _{i for} correction writing is at most several percent of the irradiation density Q _o for pattern writing. For this reason, even if the correction dose is selected from, for example, the ten discrete values, the influence of the error of the long-distance photosensitive action by this becomes 1/100 or less of the correction dose by the long-distance photosensitive action, This is because it can be ignored in practical use.

[0033]

As described above, according to the present invention, it is possible to eliminate the influence of the long-distance photosensitive action, and to obtain a highly accurate drawing pattern.

[Brief description of the drawings]

FIG. 1 is a flowchart showing a processing procedure of an embodiment of an electron beam writing method according to the present invention.

FIG. 2 is an explanatory diagram illustrating division of a drawing area according to the embodiment;

FIG. 3 is an explanatory diagram illustrating a drawing method of a divided small area.

FIG. 4 is an explanatory diagram for obtaining a continuous dose density distribution by blurring shots.

FIG. 5 is an explanatory diagram for explaining one method of obtaining an irradiation dose density for correction drawing.

FIG. 6 is a perspective view showing a configuration of an electron beam writing apparatus.

FIG. 7 is an explanatory diagram for explaining a problem of a conventional electron beam drawing method.

[Explanation of symbols]

 Reference Signs List 10 sample 20 drawing area 25 small area 28 shot area 52 electron gun 54 focusing lens 56, 62 forming aperture 58 projection lens 60 forming deflector 64 high-speed sub deflector 66 objective lens 68 high-precision main deflector 80 sample

Continuing from the front page (72) Inventor Mitsuko Shimizu 1 Toshiba-cho, Komukai-shi, Saiwai-ku, Kawasaki-shi, Kanagawa Prefecture Inside the Toshiba R & D Center (72) Inventor Susumu Oki 1 share of Toshiba-cho, Komukai-shi, Kochi-ku, Kawasaki-shi, Kanagawa Inside the Toshiba R & D Center (72) Hiroto Yasue, Inventor 1 Komukai Toshiba-cho, Saiwai-ku, Kawasaki-shi, Kanagawa Prefecture Inside the Toshiba R & D Center (72) Inventor Higurashi, etc.Toshiba Komukai-Sachi-ku, Kawasaki-shi, Kanagawa Town 1 Toshiba R & D Center (72) Inventor Takayuki Abe Komukai Toshiba 1 Co., Ltd. Toshiba R & D Center (72) Inventor Jun Takamatsu Kawasaki City, Kanagawa Prefecture Komukai Toshiba-cho 1 Toshiba R & D Center (72) Inventor Hitoshi Sunashi Oshimu Toshiba-cho 1 Koyuki Toshiba-cho, Kawasaki-shi, Kanagawa Prefecture Toshiba R & D Center (72) Inventor Naoji Shimomura Kawasaki, Kanagawa Prefecture 1 Komukai Toshiba-cho, Ichiko-ku Toshiba R & D Center Inside (72) Inventor Yoshiaki Hattori 2068-3 Ooka, Numazu-shi, Shizuoka Toshiba Machine Co., Ltd. Numazu Office (72) Inventor Tomohiro Iijima 2068-3, Ooka, Numazu-shi, Shizuoka Numazu In-house (72) Inventor Takashi Saito 2068-3 Ooka, Numazu-shi, Shizuoka Toshiba Machine Co., Ltd. In-house Numazu (56) References JP-A-63-58829 (JP, A) JP-A-1-270317 (JP) JP-A-6-21215 (JP, A) JP-A-10-10701 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H01L 21/027 G03F 7/20 504

Claims (1)

(57) [Claims] 1. An irradiation density distribution ρ () per unit irradiation amount due to a long-distance exposure at a point X in a writing region when a point Y in a writing region of a sample is irradiated with an electron beam of a unit irradiation amount. X, Y); and a step of dividing the drawing area of the resist-coated sample, on which the pattern is to be drawn, into n (≧ 2) small areas i (1 ≦ i ≦ n). determining a dose D _i in each small area i based on to the pattern, the dose density distribution ρ (X, Y) and the dose D _i
Based on the steps of obtaining a dose density F _i by long-distance photosensitive action in each small area i, based on the dose density F _i at each of the small areas i, of the corresponding region, the effect of long-distance photosensitive action determining a corrected dose density C _i for removal, the correction dose density C _i obtained for each small area i, a step of irradiating an electron beam to fill the corresponding small regions, the drawing data based comprising the steps of: performing a pattern writing step of obtaining the corrected dose density C _i in each small area i is corrected dose B _i for removing the influence of long-distance photosensitive action in each small region i _Is determined so that F _i + B _i = 0, and an offset dose B _os is introduced, and B
determining the offset dose B _os so that the value of _i + B _os is non-negative; and determining the corrected dose density C _i as C _i = B _i + B _os. Electron beam drawing method.