JPS6386432A - Electron beam lithography method - Google Patents

Abstract

PURPOSE:To perform lithography in a highly accurate manner by a method wherein, when the beam size of a variable molding beam is controlled, (beam current)/(beam area) is kept constant in the combination of all the beam sizes. CONSTITUTION:The voltage, with which the maximum beam size will be obtained, is applied to a deflector 18, and the beam size at that time is measured. The voltage is regulated until the measured value becomes equal to the maximum size. When the beam size is measured, fine particles 28 of gold and the like are arranged on a sample surface 17, and the reflected electron signal is detected when the beam is scanned on the fine particles 28. When beam sizes other than the largest size are decided, they can be adjusted using a signal P giving (beam size)/(voltage change)=gain G and a signal Q giving an offset-value. In that case, a Faraday cup 21 is moved to the point located directly under a lens-barrel, the beam current when the beam size is changed is measured, and the beam size is adjusted by using the measured value. As a result, a highly accurate lithography can be performed even for a pattern formed of a number of microscopic beam sizes.

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial Application Field) The present invention relates to an electron beam lithography method using a variable shaped beam, and in particular to an electron beam lithography method that aims at high accuracy in setting beam dimensions. Regarding the method.

(Prior art) Conventionally, when controlling the beam size in an electron beam lithography system using a variable shaped beam, the beam size is actually measured when voltages under two conditions are applied to a deflector that changes the beam size. , a method has been adopted in which a voltage that provides an arbitrary beam size is calculated from the two voltage conditions and the measured value of the beam size. In this case, there is a problem in measuring the beam size with a beam size smaller than the beam edge resolution, so a beam size that is sufficiently larger than the beam edge resolution has to be selected, and with small beam sizes, it becomes an extrapolated value and the error becomes large. there were.

In order to avoid this problem, a method has been proposed in which the beam current is measured instead of the beam size and a voltage that provides an arbitrary beam size is calculated from the relationship between the voltage conditions and the measured value of the beam current. This is based on the fact that beam current can be measured with high accuracy no matter what the beam size is.

However, as a result of actually trying to apply this method to drawing, it became clear that there were the following problems.

In other words, a pattern drawn with a relatively large beam has high precision, but a pattern drawn with a small beam while shifting the beam position does not necessarily have high precision, and in some cases may be less precise than the design grow bigger,
In other cases, it became clear that the dimensions were smaller than the design dimensions and could not be controlled properly. This is because, for example, when the beam size in the X direction is changed, the beam size in the y direction also changes slightly, and there is a slight error in beam current measurement. It turns out that this is because a proportional relationship with the area no longer holds true.

(Problems to be Solved by the Invention) As described above, in the conventional method, there is an error between the specified value of the beam dimension and the actual beam dimension, and in particular, there is an error with the design dimension in a pattern drawn with a large number of micro-dimensional beams. The problem was that it was large.

The present invention has been made in consideration of the above circumstances, and its purpose is to be able to draw with high precision even a pattern formed with a large number of minute beam dimensions, and to improve overall drawing accuracy. An object of the present invention is to provide an electron beam lithography method that can be improved.

[Structure of the Invention] (Means for Solving the Problems) The gist of the present invention is that the dose per unit area of the pattern is
The object is to make the same effect when using a beam of large size and when using a beam of small size.

One of the requirements for maintaining the accuracy of the drawn pattern is that the dose per unit area of the pattern is the same when using a large beam size and a small beam size. In the case of the conventional method, the beam current error was the same for both small and large beam sizes.

Therefore, the dose error becomes relatively large as the beam size becomes smaller, and the precision of patterns drawn with minute dimensions is unstable. Therefore, it is sufficient to convert the measured beam current into a dose by dividing it by the beam size on the data, and perform 1lltllll so that this value is equal for various beam sizes.

The present invention focuses on such points, and in an electron beam lithography method in which a desired pattern is drawn using a variable shaped beam, when controlling the beam size, the beam current is adjusted in advance at specified values of the beam size under at least three conditions. are measured, and the beam N current is divided by the beam area on the data determined from the beam size specification value, so that the value of (beam current)/(beam area) n Q<n≦1 is within the allowable range. This is a method in which parameters for beam size control are controlled.

(Function) With the above method, the dose per unit area of the pattern is approximately the same whether a beam with a large size is used or a beam with a small size is used.

Therefore, it is possible to draw a pattern according to the designed dimensions, regardless of the beam dimensions.

In fact, 0.05.0.1, 0.25.0.5 [μ
When a pattern of 0.5 [μ7FL] was formed by scanning a beam of 71L], there was a maximum variation of 0.2 [μTrL] when the conventional method of beam control was performed. On the other hand, by using the method of the present invention, the difference can be reduced to 0.05
It was possible to make it smaller than [μTrL].

(Example) Hereinafter, details of the present invention will be explained with reference to illustrated embodiments.

FIG. 1 is a schematic configuration diagram showing an electron beam lithography apparatus used in an embodiment of the method of the present invention. In the figure, reference numeral 11 denotes an electron gun, and an electron beam emitted from the electron gun 11 is focused by a condenser lens 12 and irradiated onto a first beam shaping aperture mask 13 . Aperture mask 1
The aperture image No. 3 is formed by the projection lens 14 onto the second beam shaping aperture mask 15 . and,
The aperture image of the second beam shaping aperture mask 15 is formed and irradiated onto the sample surface 17 by the objective lens 16.

A beam size variable deflector 18 is arranged between the first aperture mask 13 and the projection lens 14, and by deflecting the beam with this deflector 18, the first The aperture image position changes, and the size of the beam imaged onto the sample surface 17 changes. Furthermore, a beam scanning deflector 19 is arranged below the objective lens 16.
9, the dimension-controlled beam is scanned over the sample surface 17.

Further, 20 in the figure is a CPU, and the amount of beam current detected by the Faraday cup 21 is supplied to the CPU 20 via a voltage amplifier 22 and a digital voltmeter 23. In the CPU 20, a deflection signal is obtained based on the input information, and this deflection signal is sent to the D/A converter (
(hereinafter abbreviated as DAC) are sent out on 24 and 25, respectively.

The signal D/A converted by the DAC 24 is applied to one electrode of the beam size variable deflector 18 via an amplifier 26. Furthermore, the signal D/A converted by the DAC 25 is transmitted to the beam size variable deflector 1 via the amplifier 27.
8 is applied to the other voltage. Then, the voltage applied to the deflector 18 causes the aperture 15 to open as shown in FIG.
A and the aperture 13a (the overlapping portion with 113a', the shaded area in the figure) are variable, thereby making it possible to change the beam dimension. In this example, the dimensions of the variable beam varied in the X direction and remained constant in the y direction.

Next, a beam size setting method using the above device will be explained. First, a voltage that gives the maximum dimension is applied to the deflector 18, and the beam dimension at this time is measured. The voltage is then adjusted until this value equals the maximum dimension. Here, in order to measure the beam size, fine particles 28 of gold or the like are placed on the sample surface 17, and a reflected electron signal when the beam is scanned over the fine particles 28 is detected.

Furthermore, when determining a beam having a dimension other than the maximum dimension, other beam dimensions are also adjusted using a signal P that gives (beam size)/(voltage change) - gain G and a signal Q that gives an offset value. In that case, the Faraday cup 21 is moved directly below the lens barrel, and the beam size is adjusted using the measured value of the beam current amount when the beam size is changed.

In FIG. 3, the actual value 11A is the relationship between the designated value of the beam dimension and the amount of beam current. If the beam dimension specification value and the beam current are in a proportional relationship, it can be said that they have been adjusted approximately correctly. However, the results of measuring the dimensions of patterns formed with various beam dimensions under these conditions are shown in Figure 4.
! Indicated by D. From this figure, we can see that the dimensions differ depending on the type of beam that divides a certain line width.
This becomes a factor that hinders high-precision drawing. Note that this example shows the punching dimensions when a pattern of <0.8 [μTrL] is drawn using 1 to 8 beams, as shown in FIG. 5.

In order to investigate the cause of this dimensional variation, the current density distribution was calculated by dividing the data of Actual 11A in FIG. 3 by each beam dimension. The results are shown in broken NIB in FIG. B
As is clear from the comparison between and D, the pattern drawn with a larger current density has a thicker pattern size (,I!
It was found that patterns written with a beam with a lower current density have smaller dimensions.

Considering the above experimental facts, the software for beam size determination was changed so that the dose accuracy was within 1% for all beam sizes. The method was substantially the same as the method described above, and the focusing conditions were such that the dose deviation in each dimension was minimized. The result of adjusting the beam size using this method is shown by the dashed line C in FIG. From this, it can be seen that the dose deviation is within 1%. In addition, the results of measuring the dimensions of the pattern drawn under these conditions are shown in curve E in Figure 4.
Indicated by As is clear from comparing D and E, the conventional method caused large pattern dimensional errors (especially when using a beam with a small diameter), whereas the method of this embodiment It can be seen that the pattern dimensions drawn using the beam dimensions are equal.

As described above, according to the method of this embodiment, by changing the evaluation function from a proportional relationship between beam current and beam size to a constant current density relationship, no matter what size beams the graphic pattern is divided into, Allows you to draw patterns with the same dimensions. This is thought to be because the dose given to the resist is constant. Therefore, it is possible to reduce the pattern size error due to the difference in beam size, which is a problem when using a variable shaped beam, and it is possible to improve the writing accuracy.

Note that the present invention is not limited to the method of the embodiment described above. For example, the beam dimensions are not limited to the X direction,
The beam size may also be varied in the X direction.

In this case, instead of the specified beam size value for dividing the beam current, (specified beam size value in the X direction) x (specified beam size value in the X direction), that is, the beam area on the data may be used. In addition, in the example, in order to keep the dose constant, ilJ was controlled so that the value of (beam current)/(beam area) was within the allowable range for all beam dimensions, but this control is not necessarily performed for all beam dimensions. This is not necessary, and it is generally sufficient to use at least three beam dimensions (large, medium, and small).

Furthermore, in order to avoid the need to control the precision of the beam current at a small beam area with higher precision than necessary, (beam current)/(beam area) n Q<n<1 may be used as the evaluation function. In this case, it becomes possible to converge the automatic program in a short time. In addition, various modifications can be made without departing from the gist of the present invention.

[Effects of the Invention] As detailed above, according to the present invention, when controlling the beam dimensions of a variable shaped beam, (beam current)/(beam area) is made constant for all combinations of beam dimensions. By this, the dose of the drawn pattern to the resist can be kept constant. Therefore, no matter what combination of beam dimensions the pattern is drawn with, it will be developed to the same dimensions, making it possible to form a pattern with high precision.

[Brief explanation of the drawing]

Fig. 1 is a schematic diagram showing the configuration of an electron beam lithography system used in a method according to an embodiment of the present invention, Fig. 2 is a schematic diagram showing the configuration of the main parts of the above equipment, and Fig. 3 shows beam dimensions and beam current. FIG. 4 is a characteristic diagram showing the relationship between beam size and pattern cutout size, and FIG. 5 is a schematic diagram showing an example of dividing a pattern to be drawn. 11...Electron gun, 12,14.16...Electron lens, 13.15...Beam shaping aperture mask, 1
7... Sample surface, 18... Beam dimension variable deflector,
19... Beam scanning deflector, 20... CPU, 2
1...Faraday cup, 28...fine particles.

Claims (3)

[Claims] (1) In an electron beam lithography method in which a desired pattern is drawn using a variable shaped beam, when controlling the beam size, the beam current is measured in advance at specified values of the beam size under at least three conditions, and the beam current The beam size control parameters are controlled so that the value of (beam current)/(beam area)^n0<n≦1, which is calculated by dividing the beam size by the beam area on the data determined from the specified beam size value above, is within the allowable range. An electron beam lithography method characterized by: (2) The electron beam lithography method according to claim 1, wherein the dimension of the variable shaped beam is varied in at least one of the x direction and the y direction. (3) When the dimension of the variable shaping beam is varied in only one of the x and y directions, the measured beam current is divided by the specified value of the beam dimension in the variable shape beam (beam current)/(beam dimension) 2. The electron beam lithography method according to claim 1, wherein parameters for beam size control are controlled so that the value of (designated value) is within a permissible range.