JPS63146332A - Charged beam focusing device - Google Patents

Abstract

PURPOSE:To perform the measurement of astigmatism and its compensation automatically and focusing simultaneously, by installing a device varying a focal position of charged beams, a device for surface scanning, a device to be integrated over a scanning surface and a device storing the integrated value in memory, altogether. CONSTITUTION:A central control circuit 10 transmits a signal to an objective lens power source 20, and sets an output current out of this power source 20 to the specified initial value. In consequence, an electron beam 1 forms a focus in the rear of a target 4, changing the focal position. Next, the circuit 10 transmits an angle of a scanning direction 0 deg. to a rotary circuit 19, while it transmits a start signal to a scan circuit 18. In consequence, the circuit 18 starts surface scanning, transmitting each of scanning waveforms X and Y to the circuit 19, while this circuit 19 transmits these waveforms X and Y, whose scanning directions are rotated, to scanning coils 26X and 26Y in accordance with the angle signal, and the electron beam is subjected to the surface scanning in the 0 deg. direction in the target 4. The signal electron 5 produced out of the target 4 is detected (6), and the detected signal is differentiated (11). This differentiated value is integrated during a period of the surface scanning, and the integrated value is stored in a memory circuit 13l.

Description

DETAILED DESCRIPTION OF THE INVENTION A. Field of Industrial Application The present invention relates to the measurement and correction of astigmatism in scanning electron microscopes, electron lithography devices, and ion microprobe devices, as well as to the charging and focusing techniques. , regarding the device.

11. Prior Art In scanning electron microscopes, electron lithography devices, ion microprobe devices, etc., a charged beam is narrowly focused by a focusing lens and irradiated onto a target, but the focusing lens generally has astigmatism, in which case the beam becomes oval and cannot be squeezed into a sufficiently thin and circular shape. FIG. 1 is a diagram showing a state in which a beam is focused along the optical axis Z when there is astigmatism, and the position along the optical axis Z is called the on-axis position.

ld is the circle of least confusion, and this position is compensated by astigmatism iE
corresponds to the position of the positive focus when A focal line 1c and a focal line 1e are formed before and after this position, and the distance between the IC and 1e is the magnitude of astigmatism.

X and Y are orthogonal coordinate axes, and the direction of the focal line IC with respect to the X axis is the direction of astigmatism.

In order to correct this astigmatism, a method of correcting astigmatism by attaching an astigmatism correction device to the focusing lens is widely used, but in the past, adjustment of the astigmatism correction device was performed exclusively by humans. . In other words, when astigmatism exists, the image droplet rotates by 90'' in the scanned image before and after the positive focus of the charged beam (so-called underfocus image, overfocus image), so this flow must be carefully controlled. Observe and precisely adjust the astigmatism correction device so that the coolness disappears.

However, the coolness of the statue is subtle, so whether it exists or not,
It is very difficult to judge the direction, amount, etc., and requires a great deal of skill and effort from the operator, which is a big problem for general users.

In order to solve the above problem, the present inventor previously made the following invention to automate the measurement and correction of astigmatism (
(Japanese Patent Application No. 57-128136).

FIG. 2 is a diagram showing the principle of the above invention. Reference numeral 1 shows a cross section of a charged beam, such as an electron beam, whose optical axis is perpendicular to the plane of the paper, and the beam is focused by a focusing lens (not shown) and irradiates the plane of the paper. Since the electron beam 1 has astigmatism, it has an elliptical shape, and its diameter in one direction is DI.

If the focusing lens is, for example, an electromagnetic lens, the focal length is changed by changing its excitation current, and the excitation current when Dl becomes the minimum value is set to 1. Next, regarding the direction m rotated by a certain angle from direction 1, for example, 45'', similarly, set the excitation current of the focusing lens that gives the minimum diameter of the electron beam 1 in the m direction to 1, and further rotate the n direction by 45''. The excitation current for is set to 17. The current 1 obtained in this way
1゜1,. 1. From this, the direction λ and magnitude N of astigmatism can be determined by the following equations.

N=KI ((III −1,) 2 + (
1m-1! )2) l/2. , (2) + here
41 is a constant specific to the focusing lens.

In addition, if the astigmatism correction device is, for example, a hepolar electromagnetic type, the correction current is 1. , I, is determined by the following equation.

Ix=2([(In-1m+) (L-1+)]c
os2φ1, ==2 ([(L-L) (L
It)]cos2φ[(ill-1111) +(11
111+)]s+n2φ), (4) where 2 is a constant determined by the focusing lens and the astigmatism correction device. Further, the lens electric current iQ I o corresponding to the positive focal point position after astigmatism correction is given by the following equation.

io = (1+ +In)/2, (5) Therefore, if we find the lens electrolyte that gives the minimum diameter in at least three different directions, we get:
Astigmatism can be measured using equations (1) and (2), and (3),
It can be corrected using equation (4).

Furthermore, since focusing can be performed at the same time using equation (5), there is no need to perform a separate focusing operation on the astigmatism correction device.

To obtain the above lens current, use the rc circuit as shown in 2 in Figure 2.
Using a target with continuous end faces, I, m.

The lens electrolyte that provides the minimum beam diameter can be measured by linearly scanning the lens in three n directions, changing the lens current. Further, the minimum beam diameter may be determined by, for example, differentiating the signal waveform of secondary electrons generated from the target and obtaining the maximum differential value. Note that any number of scanning directions may be used as long as there are three or more directions, and the mutual angle difference is also 4.
It may be other than 5'', and in either case, a solution to the equation exists, and it is sufficient to slightly change equations (1) to (5).

However, although this method can be applied to special targets with end faces such as 2 in Figure 2, it cannot be applied to targets with arbitrary shapes such as general samples such as those used in scanning electron microscopes. However, astigmatism could not be accurately measured or corrected, and focusing could not be performed correctly, and the cause was unknown.

OBJECT OF THE INVENTION The present invention solves the drawbacks of the conventional two-way one-piece astigmatism correction device, and can be used for general targets with arbitrary shapes such as specimens in scanning electron microscopes, etc. The object of the present invention is to provide a charged beam focusing device that automatically measures and corrects astigmatism and also performs focusing at the same time.

20 Structure of the Invention The present invention gradually changes the focal position of the charged beam to the front and back of the target surface along the optical axis, and focuses the charged beam at three or more points per target.
- scan the surface in different directions, and generate a signal based on the change in the intensity of the signal generated from the target by the charged beam irradiation over the scanning surface! A, the integrated value is stored for each scanning direction in correspondence with the axial position of the focal point, and the axial position is calculated for each scanning direction using the value corresponding to the axial position that includes the two memorized focal lines. The invention relates to a charged beam focusing device that calculates an eigenvalue of , and uses the eigenvalue to measure, correct, and focus astigmatism.

E. Example The present inventor investigated the reason why the prior art described in the prior art section cannot be applied to general targets, and found that the reason lies in the following points. This will be explained below with reference to FIG.

In the figure, 3 a e J t) y JC is a latex sphere generally used as a target for astigmatism correction. These spheres are distributed at irregular positions, so when the beam is line-scanned in the I, m, and n directions, the beam may accidentally cross the end face of the sphere at right angles, as in 3C.
In general, it crosses diagonally as shown in 3a and 3b, and the angle at which it crosses varies. If the beam crosses the end face diagonally,
The focal positions giving the minimum beam diameter will be different from those in the case of traversing at right angles, and they will be distributed between the two focal lines according to the traversing angle. In this case, even if you differentiate the signal waveform of secondary electrons etc. generated from the target and find the point where the differential value is maximum, it will not give the focal position where the beam diameter is minimum in the scanning direction, and therefore Point (1)
Even if substituted into equations (5), it is not possible to obtain a signal related to astigmatism measurement, correction 1E, and focusing.

To solve the problems with such common targets,
First of all, it is necessary to make the beam scan a two-dimensional surface scan. As a result, the uniqueness of the angle of the target end face that happens to be located on the scanning line "2" is averaged out, and even if the scanning direction is rotated, approximately the same location on the target is scanned by changing only the direction. Errors caused by scanning different locations can be prevented. Second, as mentioned above, the beam crosses the target at various angles, and the focal position where the beam diameter is minimum differs depending on the angle, and these are distributed between the two focal lines, so there is a differential value at +1t. Rather than detecting a single point where the maximum value occurs, it detects signals related to the diameter over a wide range of axial positions including the two focal lines, and detects signals related to the diameter within that range. It is necessary to calculate a signal related to astigmatism using the signal f.

The present invention is based on the investigation and discovery described above,
... Automatically measures or compensates for astigmatism even in a ship-shaped target, and at the same time enables focusing. The following description will be made according to the embodiment shown in FIG.

Reference numeral 1 in FIG. 4 indicates an electric r-beam emitted from an electron gun (not shown), which has astigmatism, and is focused by an objective lens 27 to irradiate the target 4.

The target may be a general test sample or a special target for astigmatism correction. Also, 26X and 26Y are scanning: 1 ill, and the electron beam 1 is two-dimensionally scanned over the target. 25
X*25Y is an astigmatism correction coil. 5 is electronic IK!
These are signal electrons such as secondary electrons and reflected electrons generated from the target 4 by radiation, and are detected by the detector 6.

In this device, astigmatism measurement, correction, and focusing are performed in the following steps. First, the central control circuit 10 sends a signal to the objective lens power supply 20 to set the output current from the power supply 20 to a predetermined initial value. As a result, the electron beam 1 is focused behind the target 4 (downward in the figure) and becomes underfocused beyond the focal line position.

Next, the central control circuit 10 sends an angle signal of O° in the scanning direction to the rotation circuit 19, and also sends a start signal to the scanning circuit 18. As a result, the scanning circuit 18 starts surface scanning and sends the scanning waveforms x and Y to the rotation circuit 19, and the circuit 19 sends the scanning waveforms X and Y whose scanning direction has been rotated according to the angle signal to the scanning coils 26X and 26Y. The electron beam is surface-scanned on the target 4 in the 0° direction.

Signal electrons 5 generated from the target 4 are detected by a detector 6, and the detection signal is differentiated by a differentiation circuit 11. This differential value is calculated by the integrating circuit 12 during one surface scan
The calculated and integrated value (hereinafter referred to as an integrated differential value) is stored in the storage circuit 131.

Next, the central control circuit 10 sends a signal to the objective lens power supply 20 to set the output current from the power supply 20 to a value of 1, which is the initial value ' plus a certain increment, and further sends a signal to the scanning circuit 18 to set the output current from the power supply 20 to a value of 2 [i
jl [Send 1 start signal. As a result, electrons and -)
, is surface scanned with the focal length slightly shorter than the first time, and at that time! A37. The differential value is stored in the storage circuit 131. Below, the action of the same ladder is as follows: the beam is forward (in the figure)
2 directions) is repeated until an overfocus state that exceeds the focal line position is reached, and a series of 1
The AT1 differential value is stored in the storage circuit 13+. Furthermore, the central control circuit lO performs the above operation in the scanning direction of 45° and 90°.
Performed against. As a result of these operations, the scanning direction O°, 4
The integrated differential values at 5° and 90° are stored in the memory circuit 131p 13 in correspondence with the on-axis position in the range including the two focal lines.
m p 13 n respectively. Normally, the objective lens current value is stored instead of the axial position, but the axial position has a simple proportional relationship with the lens current value within a narrow range, so the axial position is actually stored. become. Further, since the objective lens current is normally changed by a constant increment, it is sufficient to memorize the starting value, the increment, and the number of increments without storing each current value.

The function curves of the integrated differential value with respect to the on-axis position obtained in this way are 301 (in the case of scanning direction 0°) and 30 in Figure 5.
m (45°) and 30n (90°).

To derive accurate information about astigmatism from this data, as mentioned above, it is necessary to use the entire range of function values. Although several methods are possible for implementing this, the first example is a method using the center of gravity of the function curve, and the function analysis circuit 14 in FIG.
3m and 13n, the geometric center of gravity with respect to the axial position direction of the function curves stored in 3m and 13n is calculated. a2tt 32 in Figure 5
mt; 32n is the position of the center of gravity obtained in this way, and the area under each curve is equal on the left and right of each center of gravity position. The astigmatism calculation circuit 15 in FIG. 4 calculates (1), (
2) Calculate the signal representing astigmatism by substituting into the equation, (3)
, a signal for compensating for astigmatism is calculated by substituting into equation (4), and a signal for focusing □ is calculated using equation (5). However, for the constants Kl and 2 in these equations, it is necessary to use values specific to the method using the center of gravity, and these must be determined in advance in a separate experiment.

The calculated signal is sent to the central f111a1 circuit 10,
A signal representing astigmatism is sent from the control circuit 10 to the display device 16 to display the astigmatism, and also to compensate for the aberration 11-. The signals are sent to the astigmatism correction power supplies 17X and 17Y, and the correction power supplies 17X and 17Y convert the signals into currents and send them to the astigmatism correction illuminations 25X and 25Y to correct the astigmatism of the electron beam l. Further, a focusing signal is sent from the central control circuit 10 to an objective lens power source 20, and a focusing current is sent from the power source 20 to the objective lens 27 to perform focusing.

In Figure US 6, the function curves of the l7tTI differential value were obtained again after the astigmatism correction, and the positions of the three curves in the left and right directions are the same, indicating that the astigmatism has been completely corrected. Further, 40 in the figure is the position of the positive focal point calculated by equation (5), which coincides with the maximum position of the three curves, indicating that focusing is performed correctly.

Next, a second example of functional analysis will be explained with reference to FIG. 301.30m and 30n in the figure are function curves of integrated differential values, as in FIG. In this method, first, the positions of the center of gravity Pt321, 32n are determined using the method described above, and the positive focal point position 4o is determined using equation (5). Next, the right surface f/l L of the function curve 301 divided by the positive focus position,
Find the ratio of /L to the area L 1 on the left side, and similarly calculate Mr/M+ for the function curve 30 m p 30 n.
, Nr/N+ (not shown). The sign of these numerical values is positive when the area on the right side is larger than the area on the left side, and negative when it is smaller.

In this way, /I, +, Mr/M+ −
Assuming the values of N, /N, to be I+, I-, I, (1) to (
5) Substitute into the equation to calculate a signal representing astigmatism, a signal for correcting astigmatism, and a signal for focusing. However, since the constants Kl and 2 in the above equation are values specific to this embodiment, they are determined in advance in a separate experiment.The operation after calculating the signal is the same as in the previously described embodiments.

In yet another embodiment, first the positive focal point position is determined in the same manner as in the previously described embodiment, and then the slopes of the three function curves at the positive focal point position are determined. Each of these gradients is 1. ．． 1. ,
In is substituted into equations (1) to (5), and the subsequent operations are the same as in the previously described embodiment. In this case too, the constant KI
, it goes without saying that 2 is a different value.

In this way, there are various methods for functional analysis, but all of them use the values for the axial positions in the range that includes the two focal lines of the collected data to calculate the unique values 1+, I, , , for each scanning direction. In is determined, and astigmatism and focus are measured or corrected from these eigenvalues.

Although the above embodiment describes the case where an electron beam is used, other charged beams such as an ion beam may also be used, and the signal for determining the beam diameter may be not only secondary electrons but also reflected electrons. , secondary ions, X-rays, target absorption current,
It is also possible to use signals such as the internal electromotive force of the target,
Furthermore, as means for focusing, scanning, and astigmatism correction of the beam, an electrostatic method may be used instead of an electromagnetic method.

Also, the signals to be integrated are not limited to differential values; for example, signals based on changes in signal intensity generated from the target due to scanning, such as integrating the peak value of the signal or rectifying the alternating current component of the signal, can be used. Furthermore, in the above embodiment, the focus position of the charged beam is changed by a focusing lens, but a so-called dynamic focus coil may be provided separately from the focusing lens, and the focus position may be changed by this.
If you only want to measure astigmatism, you only need to display the calculation results on the display VI. Conversely, if you only want to correct astigmatism, you can supply the calculation results to the correction coil via the correction power supply. If only focusing is desired, it is sufficient to simply supply the calculation result to the objective lens via the objective lens power supply. Also, the scanning direction of the beam is 3 or more,
For example, four directions may be used.

Furthermore, based on the invention described above, it is possible to further enhance the effect of the invention by partially incorporating or changing the invention. In the above embodiment, the astigmatism is measured and corrected immediately after the start of the operation, but before that, it is recommended to perform the 11 am adjustment of the 1-move focusing.
After the start of the operation, first, data of the function curve of the integrated differential value is collected in an arbitrary scanning direction, for example, in the 0° direction, using the same procedure as described above. In this case, the range in which the focal position is changed is widened to ensure that the target plane exists within that range. Next, the position of the center of gravity of the obtained function curve is determined, and in the subsequent astigmatism measurement and compensation iE operations, the focal point position is changed around the center of gravity position determined above. . Although this center position does not necessarily match the position of the positive focus,
Since it is certain that the focal point is in the vicinity, if the focal position is changed around this point, it is possible to include the two focal lines within the smallest change range, thus shortening the operation time and improving reliability. I can do it.

In addition, in the embodiments so far, examples have been described in which astigmatism is corrected once, but the present invention actively utilizes the feature that it can also measure astigmatism at the same time, and performs correction multiple times. It is also possible to increase the accuracy. That is, when astigmatism is measured and corrected for the first time, the magnitude of the measured astigmatism is compared with a preset value, and if the measured value is larger than the set value, two more measurements are performed. If you repeat the same procedure until the measured value becomes smaller than the set value, you can reliably reduce the remaining astigmatism to an extremely small value, and the astigmatism correction will become even more perfect. .

Also, when performing astigmatism correction multiple times as described above, the residual astigmatism becomes smaller each time, so
The distance between the two focal lines also becomes narrower, so the range in which the focal position of the charged beam is changed to obtain the function curve can also be gradually reduced each time.In this way, the operating time can be reduced. The practical value increases.

Furthermore, in history, the above-mentioned astigmatism measurement was carried out at a certain JIJ.
The measurement results are memorized at regular time intervals over the period I, and the increasing tendency of astigmatism can be determined from the changes in the memorized measured values over time. A gradual increase in astigmatism is usually caused by contamination inside the beam focusing device (although a sudden increase may be due to a malfunction in the electrical system, etc.). , it is possible to obtain information about the status of equipment when it is malfunctioning, and from this it is also possible to judge when to perform maintenance work such as cleaning and repair of the equipment, which is of great practical value. It goes without saying that it is convenient to automatically measure and store the astigmatism at regular time intervals.

1. Effects of the Invention 2. As explained above, the present invention gradually changes the focal position of the charged beam along the optical axis before and after the target surface, so that the charged beam can be directed in three or more different directions on the target. A signal based on a change in the signal intensity generated from the target due to the scanning by the charged beam irradiation is transmitted across the scanning plane at f7I$?
:L, f7tTI values are stored for each scanning direction in correspondence with the axial position of the focal point, and the axial position is determined for each scanning direction using the values corresponding to the axial position including the two stored focal lines. Calculate the eigenvalues, and use the above eigenvalues to measure astigmatism, correct astigmatism, and focus (to calculate the signal necessary for either one, the direction and magnitude of astigmatism It is possible to quantitatively and accurately measure the position of the positive focal point, and it is also possible to automatically correct astigmatism and focus.Furthermore, based on the above-mentioned procedure, by partially recombining or changing it, Furthermore, it is possible to enhance the effects of the invention.As a result, the operator can easily handle the charged beam device without requiring skill or patience, and the practical effects are extremely large.

[Brief explanation of the drawing]

Fig. 1 is a diagram for explaining astigmatism, Fig. 2 is a diagram for explaining the conventional technology, Fig. 3 is a diagram for explaining the cause of the drawbacks of the conventional technology, and Fig. 4 is a diagram for explaining the present invention. 5 and 6 are function curve diagrams for explaining the operation of the above embodiment. FIG. 7 is a diagram for explaining the operation of another embodiment of the present invention. It is a diagram of a function curve for. In addition, in the symbols shown in the drawings, 1: electron beam, ld: circle of least confusion, lc, lc: focal line, 2: target with a straight end surface, 3a. H3be 3 c; Radex sphere, 4; Target, 5; Signal electron, 6; Detector, 10: Central control circuit, 11; Differential circuit, 12: Integration circuit, 13 I, 13m, 1; 3
n: Memory circuit, 14; Function analysis circuit, 15; Astigmatism calculation circuit, 16; Display device, 17X, 17Y; Astigmatism compensation iE power supply, 18; Scanning circuit, 19; Rotation circuit, 20; Objective lens power supply, 25X , 25Y; Astigmatism iE J 4 k, 26
X, 26Y; scanning 4np, 27; objective lens, 301
．． 30m, 30n; Function curve, 321.32m, 32
n: position of the center of gravity of the function curve, 40: position of the positive focus
It is. Figure 1 S Figure 61 Calculation q1 Figure j2J! 32M

Claims (1)

[Scope of Claims] 1. Means for gradually changing the focal position of the charged beam to the front and back of the target surface along the optical axis, means for scanning the surface of the target with the charged beam in three or more different directions, and a charged beam. means for integrating a signal based on a change in signal intensity generated from the target due to the scanning over the scanning plane; and storing the integrated value for each scanning direction in correspondence with the axial position of the focal point. means for calculating an eigenvalue of an axial position for each scanning direction using the stored value corresponding to an axial position including the two focal lines; A charged beam focusing device comprising means for calculating a signal necessary for at least one of aberration correction and focusing. 2. The charging method according to claim 1, characterized in that the approximate position of the focal point is determined by coarsely adjusting the focusing of the charged beam, and then the focal position of the charged beam is changed around the approximate position. Beam focusing device. 3 When astigmatism is measured and corrected, the magnitude of the measured astigmatism is compared with a preset value, and if the measured value is larger than the set value, the astigmatism is corrected again. A charged beam focusing device according to any one of claims 1 to 2, characterized in that it is configured as follows. 4. Astigmatism correction is performed multiple times, and the range of change in the focal position of the charged beam is made smaller in subsequent corrections than in the previous correction. Charged beam focusing device. 5. Memorize the measurement results of astigmatism for a certain period of time,
5. A charged beam focusing device according to claim 1, wherein the state of the device is detected from changes in stored measured values over time.