JPS61214342A - Circuit for deflecting charged particle beam - Google Patents

Abstract

PURPOSE:To reduce the time required for drawing all fields by measuring the difference between the intended positioning level and the output level of a main deflection system and feeding the detection result back to an auxiliary deflection system thereby simultaneously operating the main deflection system and the auxiliary deflection system. CONSTITUTION:A charged electron beam deflector for an electron beam exposure device or a similar device consists of a main deflection system 10 and an auxiliary deflection system 20 the main field of each of which is divided into sub-fields. A D/A convertor 1 is used to analogize information about the position of the central sub-field of the main deflection system 10 and a detector 4 is used to measure the difference between the input and the output of an amplifier 2 to which the analogization result is given. Positional information for drawing a pattern is analogized and fed back to the auxiliary deflection system 20, thereby simultaneously performing main deflection 3 and auxiliary deflection 7. Therefore, it is possible to reduce settling time for moving the beam to the central sub-field, thereby reducing the time of drawing.

Description

DETAILED DESCRIPTION OF THE INVENTION (Technical Field of the Invention) The present invention relates to a charged particle beam deflection system used in an electron beam exposure apparatus or the like.

(Background of the Invention) The required conditions for a charged particle beam deflection system used in an electron beam exposure device or the like are (1) high resolution and high stability. (2) Large dynamic range. (3)
It must be fast. However, it is difficult to simultaneously satisfy these requirements with one deflection system, so two deflection systems, a main deflection system and a sub deflection system, are used to cover the drawing area (main field) from 100 to 100 io. A charged particle beam (hereinafter simply referred to as a beam) is deflected to the center of the subfield to be drawn using a main deflection system that satisfies the requirements of a small area of about . This is an excellent method that can satisfy the three conditions (1) to (3) at the same time by driving the beam at high speed and writing using a sub-deflection system that satisfies the condition (3).

Namely.

A drive system with a large dynamic range and high resolution of the main deflection system takes time to settle (the time it takes to reach the target value), so it should be used only when moving subfields.
Since pattern drawing within a subfield has a small dynamic range and low resolution, this method uses a sub-deflection system that can be driven at high speed to increase the drawing speed, and the total drawing time is reduced by using one deflection system. This is a method that can significantly shorten the time compared to the previous method.

However, even with this method, the main deflection system has finished drawing. The time loss required to settle the beam from one subfield to the center of the second subfield to be drawn cannot be ignored.

Especially the drawing time when there are many subfields.

This is becoming a bottleneck for further shortening.

(Objective of the Invention) The present invention solves these drawbacks by extremely shortening the time required for settling of the main deflection system for moving the beam to the center of the subfield to be drawn in the secondary drawing.
The purpose is to dramatically shorten the total drawing time.

(Summary of the Invention) The inventor of the present invention firstly discovered that, in the position of the conventional main deflection system,
All corrections related to beam deflection and positioning
If you do it with the sub-deflection system.

The main deflection system drive circuit only needs to generate a signal corresponding to the center position of each subfield, and in that case, a high-speed DAC of about 4 to 7 bits can be used as long as the accuracy and stability are good. Second, the amplifier that drives the main deflection system has a large dynamic range, so it uses a high voltage or current, and there is a limit to how high the speed can be increased. 2. By feeding back to the sub-deflection system, the settling time of the main deflection drive amplifier can be substantially shortened.
The present invention was accomplished by focusing on two points.

The present invention is configured such that corrections related to beam deflection and positioning are performed in a sub-deflection system, and furthermore, the error between the main deflection positioning information (target data) and the output signal of the main deflection amplifier is fed back to the sub-deflection system, and the error is fed back to the sub-deflection system. The technical point is to operate the sub-deflection system simultaneously with the main deflection system according to error data.

(Embodiment) FIG. 1 is a schematic diagram showing an embodiment of an electron beam exposure machine using the present invention. 11 is an electron gun, 12 is a first condenser lens, 13 is a blanking deflector, and 14 is A second condenser lens, 3 a main deflector, 7 a sub deflector, 15 an objective lens, 16 an xy stage, 10 a main deflection drive circuit, and 20 a sub deflection drive circuit. In addition to the blocks shown in FIG. 1, several correction electron optical systems are normally installed, but they are not directly related to the present invention and will therefore be omitted.

FIG. 2 shows a circuit showing an embodiment of the present invention, in which a block 10 surrounded by dotted lines is a main deflection system drive circuit 20, and a sub deflection system drive circuit 30 is a deflection electron optical system. In actual use, the electron beam must be deflected using the information in the X and Y orthogonal coordinate system, so one more set of the deflection system shown in Figure 1 is required, but the nine wheels must have exactly the same deflection system. Therefore, for the sake of brevity, only the -axis deflection system will be explained here.

1 is a main deflection D/A converter (hereinafter the D/A converter is referred to as DAC) that converts the center position information (digital signal) of the subfield of the main deflection system into an analog signal;
For example, if the main field is divided into 100×100 subfields, a minimum of 7-bit DAC is required, but it seems appropriate to use a 10- to 12-bit high-speed DAC to improve positional accuracy.

2 is a main deflector drive amplifier for driving the main deflector 3; If the main deflector 3 is an electrostatic deflector, it will have a high voltage of several hundred volts, and if it is an electromagnetic deflector, it will have a large current of several amperes. Therefore, the main deflector DAC 1
Even if a high-speed device becomes available, the overall drawing time cannot be shortened unless the settling of the main deflector drive amplifier 2 can be shortened.

Reference numeral 4 denotes an error detector for the main deflector drive amplifier, which outputs the error between the input and output of the main deflection drive amplifier 2 until it settles.

5 is a sub-deflection DAC that drives the sub-deflection system and sequentially converts the pattern position information digital signal into an analog signal for drawing the pattern, and requires a high-speed conversion function;
~12 bit high speed DAC is used.

Reference numeral 6 denotes a sub-deflector drive amplifier for driving the sub-deflector 7.

In the present invention, the dynamic range of this amplifier is 2
Although the deflection voltage output for the subfield is required, the dynamic range is several tenths of that of the main deflector drive amplifier 2, and its settling time is much shorter than that of the main deflector drive amplifier 2. Approximately several hundred ♀
sec < leprosy.

7 is a sub-deflector, and since a high-speed response is required, an electrostatic deflector is used in most cases.

8 is a correction DAC. Although omitted in the explanation so far, when actually drawing a pattern on a workpiece with an electron beam, the workpiece is fixed on an XY stage.
It is necessary to move the stage so that the center of the main field to be drawn is directly below the electron optical system. At this time, the stage does not necessarily stop at the target position, so normally the output of a stop position error detector (not shown) is fed back to the electron beam deflection system to perform a 1-equivalent correction. There is. The correction DAC 8 mainly converts this correction amount into an analog signal.

The correction DAC 8 is also used to correct the influence of an external magnetic field, distortion of the deflection system, and the like.

FIG. 3 shows the settling state of the output (a) of the main deflection DAC 1, the output (b) of the main deflector drive amplifier 2, and (c) the output of the error detector 4. As already described, the settling time TA of the main deflector drive amplifier 2 is the longest, ranging from several μsec to several hundred μsec. In conventional writing systems, when moving an electron beam from one subfield to the next, a waiting time of TA is always provided, and writing is paused for this time. In the present invention, the error detector 4 is assembled using a high-speed differential amplifier, and errors within the settling time of the main deflector drive unit 10 are corrected by the sub-deflection system. The sub-deflector drive amplifier 6 uses a high-speed amplifier in order to draw a pattern at high speed, and can sufficiently follow the output C of the error detector 4, which changes slowly on the order of several tens to hundreds of psec, and drive the main deflector. The error before settling of the drive amplifier 2 can be completely corrected, and while correcting the error of the main deflector drive amplifier 2, the pattern information input from the sub-deflection DAC 5 is also added at the input point of the amplifier, so the pattern information can be corrected at the same time. can be drawn. Figure 3(d) shows the sub-deflection DAC.
FIG. 5(e) shows the output signal of the sub-deflector drive amplifier 6 on which the output signal d of the sub-deflection DAC 5 and the settling error signal C are superimposed. As can be seen from the figure, the sub-deflector drive amplifier 6 superimposes settling errors.

However, it has the disadvantage of increasing costs because it requires a dynamic range that can deflect by two subfields, which is twice the area actually drawn.

The benefit of shortening the drawing pause (waiting time) time is much greater, reducing the pause time when moving subfields to about 1/10 to 1/100 of the conventional method, significantly improving the total throughput. I can do it.

2. In the present invention, as long as the accuracy and stability of the DAC used in the main deflection system is good, only the output corresponding to each subfield is required, so a high-speed, low-resolution DAC can be used, and an AC like this can be manufactured. things are possible. However, when using a commercially available high-speed low-resolution (about 10 to 12 bits) DAC, the accuracy is usually ±1/2LSB.
It is the rank.

The main deflection system does not meet the required positional accuracy. Therefore, commercially available D
When using AC, measure the amount of error between the target output corresponding to the center position of each subfield and the output of the main deflection DAC 2 in advance, save it in memory, and use it when directing the electron beam to each subfield. , the error amount for that subfield is read from the memory, subtracted and inputted to the correction DAC 8, and corrected, thereby making it possible to use it as a high precision DAC. If you use it like this, 1.5, 8. All DACs can be used with the same type, which is very convenient in terms of design and maintenance.

(Effect of the invention) As described above, according to the present invention, the settling time is
The error between the output of the main deflection system drive circuit and the target value is determined within the long settling time of the main deflection system, and based on this, the sub deflection systems with short settling time are operated simultaneously, so that the center of the next subfield is The settling time for moving the charged particle beam can be substantially shortened, and the writing time for the entire field can be dramatically shortened.

[Brief explanation of drawings]

FIG. 1 is a schematic diagram showing an embodiment of the present invention. FIG. 2 is a block diagram of a circuit system showing an embodiment of the present invention. FIG. 3 is a diagram of signal waveforms at various locations in the circuit system of FIG. 2. (Explanation of symbols of main parts) 1---Lifetime deflection DAC 5-111 Sub-deflection DAC 8--1 Correction DAC 2-111 Main deflector drive amplifier 4-111 Error detector 6 , ---1 sub-deflector drive amplifier 3--11-1 main deflector 7-111 sub-deflector 10--Issue deflection system drive Tabachi 20--1 sub-deflection system drive 161s 30----- Polarized electron optical system

Claims (1)

[Claims] In a charged particle deflection device having a main deflection system and a sub-deflection system, the amount of error between the positioning target value of the main deflection drive system and the output value of the main deflection drive system is detected, and the error amount is transferred to the sub-deflection drive system. A charged particle beam deflection circuit characterized in that a sub-deflection drive system operates simultaneously with a main deflection drive system upon feedback.