JPS58118114A - Drawing apparatus by charged particle ray - Google Patents

Abstract

PURPOSE:To make unnecessary the use of a transfer mechanism of a specimen stand by a method wherein a mark having a straight line with a known angle to a scanning direction is formed on the surface of the specimen and position correcting quantity is obtained from the coordinates obtained by scanning the mark and that angle. CONSTITUTION:Charged particle rays 4 focused on the surface of a specimen 1 is scanned over the whole surface of the specimen 1 by a deflector 2. On the surface of the specimen 1, there are arranged a plurality of marks having straight lines with a known angle theta to a scanning direction. A detector 5 is used to dtect a charged particle reflected from the surface of the specimen 1 when the mark passes the charged particle rays and the detected signal is supplied to a signal processor 13. The output of the processor 13 is given to a counter 15 and the number of counts given by the counter 15 is stored in a memory 17 of a computer 18. By so doing, the coordinates x of the mark is obtained by the computer 18. The shifted positional quantity DELTAx, DELTAy is obtained from the obtained coordinates and the angle theta. The shifted quantity obtained is computed by the computer 18 from the starting point txo and added to the signal given by a signal generator 16, before being supplied to the deflector 2 for correcting the deflected signal.

Description

Detailed Description of the Invention: When a desired pattern is drawn by irradiating charged particle IIM onto a charged particle material, a drawing in which the charged particle beam is scanned back and forth over the entire surface and the charged particle beam is irradiated only at the desired pattern position. The present invention relates to a correction method suitable for increasing the positional accuracy of an apparatus.

Conventional charged particle beam irradiation position correction involves moving one mark two-dimensionally and accurately measuring this moving coordinate value with a laser, while also measuring the charged particle beam and calculating the difference between the positions of the charged particle beam. As a result, I had to make corrections when drawing. Such a method requires a moving mechanism to move the mark and a measuring device to accurately measure its coordinate values.

An object of the present invention is to calculate correction values when the irradiation position of a charged particle beam deviates from the desired position without using a sample stage movement mechanism. We are committed to providing a simple method to measure # degrees.
In particular, it is an object of the present invention to provide a method in which the amount of position correction in a direction perpendicular to the above-mentioned direction can also be determined by scanning in one direction (for example, the X direction).

In order to avoid using a moving mechanism for the sample stage, a large number of reference marks with clear position coordinate values are placed on the sample surface.

Furthermore, in order to measure the positional deviation in the X direction by scanning in the X direction, a line perpendicular to the scanning direction and a line having an inclination are used as reference marks. Also, the scanning direction (10x.

Since there is a difference in the amount of positional deviation due to -x) or a positional deviation that depends on the scanning speed, the amount of positional deviation is measured and corrected using the same or very similar scanning method to that used during drawing.

An embodiment of the present invention is shown in FIG. 1. Charged particles @4 generated from a charged particle generation source are narrowly focused by a lens and imaged on a sample surface 1. The above charged particle beam 4
is scanned over the entire surface of the sample surface 1 by the deflector 2 as shown in the figure. That is, it scans in the px direction from the vicinity of one end 6 of the sample surface, and when it reaches the vicinity of another end 6, it shifts by δy and scans in the -X direction. By repeating this, the entire surface is reciprocated scanning to reach the vicinity of the other end 8 and the final end 9. On the other hand, the blanking device 3 performs control to cut the charged particle beam 4 irradiated onto the sample surface 1 . Therefore, during pattern drawing, while the charged particle beam reaches the position where the pattern exists, the charged particle lll1! is controlled by the blanking device 3. All you have to do is irradiate it.

When performing a high-speed drawing method in which the entire surface of the sample surface 1 is scanned and the charged particle beam is largely deflected, not only the irradiation position will shift due to static distortion of the deflector 2, but also the scanning direction (
The present invention recognizes the need to be able to comprehensively correct hysteretic positional deviations that depend on the scanning direction, -x direction), and positional deviations that depend on the scanning speed, such as e&field coupling of This method solves this problem and does not require any mechanism for moving the sample stage.

In order to accurately measure the irradiation position deviation, first place a sample 2 with a reference mark 21 as shown in Fig. 2 on the sample surface 1, which is the first cause.
Set to 0. FIG. 3 shows an enlarged view of this reference mark. In the same figure, the charged particle beam 30 is located at this mark 21.
Charged particles reflected from the sample surface when passing above are detected by the detector 5 shown in FIG.

The scanning method and scanning length when drawing on the sample 20 as described above,
It is assumed that when the charged particle beam 30 is scanned using a scanning method with the same speed, etc., the scanning of the charged particle beam on the sample becomes a locus 31. If there is no positional deviation, the trajectory should be 31'. In other words, the charged particles that should irradiate the center coordinate point 33' of the mark (this coordinate value has been measured with high precision and recognized by the computer 18 in advance) are located at the coordinate point 3'.
Suppose it was 3. That is, Δzl in the figure.

Δy' is the positional shift of 33'. It is difficult to measure the coordinate values of this point 33 because no signal is obtained from anywhere around the mark, but it is possible to measure the center point 34 of the mark 21 in the X direction.

This point should originally be the point 34', but by measuring the positional deviation ΔX, Δyt, the center point of the mark is 33'.
Even if there is a positional deviation in the coordinate values of Therefore, by measuring ΔX and Δy, Δx/, Δy
′ shall be the value of

Sample ii] When the charged particle beam passes over 20 as shown in 31, the detector 5 detects the reflected electrons emitted from the sample surface 20,
Suppose that the signal 35 is obtained by amplification by the amplifier 12. This signal 35 is shaped by the signal processor 13 at a predetermined slice level 36, and the rising edge and falling edge 9 of the shaped wave are taken out as trigger signals and made into a pulse signal 37. While the counter 15 is counting the clock pulses 38, the number n, to n- corresponding to the signal 37 is stored in the memory 17.

Color/removal 15 starts scanning from one end of sample surface 20 t
x0 is cleared by a clear signal at the starting point of the signal from the scanning signal generator 16. If this is done, the X coordinate value X□ of point 32 will be n, nl, and the clock pulse period t1
It is the X coordinate value of the x0 point from the scanning speed Hv, and V is the scanning speed of the charged particle beam. ). The respective X coordinate values of points 34.40 were also determined by a colleague and xs4. It is expressed as X4°. Of course, this value is the X coordinate value measured with a charged particle beam, and the known coordinate value is the value of point 33'(X□'+y31'). Based on the above values, the positional deviation amounts ΔX and Δy are determined. ΔX
It can be easily seen that ΔX=x84−xml'.

On the other hand, Δy is found using the fact that the marks are radial. That is, as shown in the figure, let the slope be θ (this is also known),
x, =x□−”U* By performing calculations, the positional deviation of each mark on the entire scanning surface can be found.In this implementation, this positional deviation is expressed as a polynomial, its coefficients are determined by the method of least squares, and this function is used to determine whether there is a mark or not. In addition, in this implementation, since the amount of positional deviation differs depending on the scanning direction (-1-x, -X direction), different corrections were performed for each direction. .

The amount of positional deviation to be corrected determined in this way ΔX, Δ
The calculator 18 calculates y from the starting point txo every moment according to the timing. An amplifier 1 converts the calculated value into a signal from a scanning signal generator 16 driven by a signal from a computer.
By correcting the circular direction signal added in step 1 and supplied to the deflector 2, deflection scanning of the charged particle beam without positional deviation becomes possible.

In the present invention, it goes without saying that the marks on the sample surface are not limited to those shown in FIG.

However, since it is necessary to find the positional deviation in the X direction only by scanning in the X direction, a pattern with an inclination may be used. On the other hand, the arrangement of the marks is not limited to that shown in FIG. 2, and the number and arrangement may be determined depending on the correction accuracy. The texture of the pattern may be different from that of the sample surface, for example, a gold pattern on chrome. Further, the correction for the obtained positional deviation is not limited to this embodiment, and it goes without saying that it is also possible to simply interpolate between marks. Also, the scanning method does not have to be exactly the same as when drawing; for example, the same thing can be done by scanning the marks lined up in a row several times and then moving to the next mark row without scanning. . However, any scanning method may be used as long as the amount of positional deviation does not differ due to the difference in scanning method from the time of drawing.

The above operations are carried out under the control of the computer 18, and the processing of the computer is as shown in FIG. In FIG. 4, each symbol is recessed as follows, and (xmttYtt) ~ (x
sve)'sy) is a known value recognized by the computer in advance.

X0=-t, x,=x,,-x,,,xsnow=x4
゜−x14Also, in the flowchart of Figure 4, the formula ■
.

(ΔX, Δy) in (2) may differ depending on the scanning direction due to the influence of hysteresis, and (ΔX.

A calculation formula for Δy) may be set.

According to the present invention, positional deviation in a direction (X direction) perpendicular to that direction can also be measured by scanning in one direction (for example, the It is effective in improving accuracy. Therefore, the entire system can be simplified, which is effective in realizing a low-cost device.

[Brief explanation of the drawing]

Figure m1 is a drawing showing a block diagram showing an embodiment of the present invention. FIG. 2 is a plan view showing an example of a sample with reference marks for measuring positional deviation. 3 is an enlarged view of a reference mark, a signal obtained during scanning of a single particle beam, and a diagram relating to the processed signal. FIG. 4 is a flowchart showing the operation of the computer 1B in FIG. DESCRIPTION OF SYMBOLS 1... Sample surface, 2... Deflector, 3... Blanking device, 13... Signal processor, 14... Oscillator, 15
...Counter, 16...Scanning signal generator, 17.
...Memo 1c18...Calculator, 21...Reference mark. 40. . 1st (2) ¥iZ[2] 1 6 LY13
薏吟圓χ 4 图 ro-,. [C7-~(1L

Claims (1)

[Scope of Claims] 1. In a charged particle beam lithography apparatus having means for narrowing a charged particle beam generated from a charged particle beam generation source and deflection means for deflecting the charged particle beam to a desired position, A sample has a plurality of marks arranged on the sample surface, each having a straight line with a known angle of inclination θ to A charged particle beam lithography apparatus having means for calculating a position correction amount of a charged particle beam from coordinates and an upper angle θ, and correcting and scanning a position of a charged particle beam using the position correction amount as a percent image. 2. The charged particle beam lithography apparatus according to claim 1, wherein the It'i nasal formula for the positive potential quantity is determined for each scanning direction. 3. The charged particle beam lithography apparatus according to claim 1, wherein a large number of the marks are arranged in a two-dimensional manner, and the mark is a 22% mark.