JPH08129985A - Scanning electron microscope - Google Patents

Abstract

(57) [Abstract] [Purpose] Smoothly and accurately move the observation field of view even if there is backlash of the stage or pitch error of the feed screw. [Structure] The sample 9 is scanned with an electron beam 3, and an enlarged image in the observation visual field on the sample 9 is displayed on a CRT display 25. The stages 12X and 12Y on which the sample 9 is mounted are
Feed screws 13X and 13Y rotated by pulse motors 14X and 14Y controlled by a micro step drive method
Driven through. A reference wafer is placed instead of the sample 9, and the stages 12X and 12Y are inverted and moved to obtain the backlash amount BX (±) and the lead screw pitch error ΔP (X) and store them in the memory 33. Incidentally, when driving the stages 12X and 12Y, the stage control means 19 corrects them.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a scanning electron microscope used for observing a fine structure of a semiconductor element or the like, and more particularly, a stage on which a sample is mounted is fed by a driving screw of a pulse motor. The present invention relates to a scanning electron microscope which is moved by a method.

[0002]

2. Description of the Related Art In a scanning electron microscope, an electron beam emitted from an electron gun and converged by a converging lens is deflected by a deflection coil, and the deflected electron beam two-dimensionally moves on the surface of a sample through an objective lens. To scan. Then, secondary electrons generated from the surface of the sample are converted into an image signal via a secondary electron detector or the like, and an enlarged image of the sample is displayed on the screen of an image display device (CRT display or the like) based on the image signal. Is displayed in.

At this time, the operator can set the observation magnification to a desired value within a predetermined range, and the scanning range of the electron beam on the sample according to the set observation magnification,
And the scanning speed and the like are adjusted. In the scanning range of the electron beam on the sample, the range specified by the operator as the observation target is the observation field of view, and an enlarged image at the set observation magnification within this observation field of view is displayed on the screen of the image display device. Is displayed.

There are two methods of moving the observation field of the sample and moving the observation image of the sample on the screen of the image display device: a mechanical field transfer method and an electrical field transfer method. . The mechanical visual field moving method is to move a stage on which a sample is placed using a movement control device, and the electrical visual field moving method is to control the scanning area of the electron beam on the sample by an electron optical system control circuit. Or the selected area of the image data within the scanning area is moved to move the observation visual field.

The electric visual field moving method has an advantage that the image can be moved at a constant speed when the observation magnification is high. On the other hand, the swing width of the electron beam is limited, and the limited width is much narrower than the moving amount of the stage. For this reason, the operator usually switches between the mechanical visual field moving method and the electrical visual field moving method as needed. Other than that, a device that automatically switches between the two types according to the observation magnification has also been used. In this regard, as a conventional stage driving means for mechanically moving the observation field of view, a stage screw is driven by a pulse motor (that is, a stepping motor) which is driven by a full step (or a half step). A feed screw method was used to send the.

[0006]

In the prior art as described above, since there is backlash of the feed screw when moving the stage, when the moving direction of the stage is reversed, the pulse motor rotates but the stage moves. There was a phenomenon in which the robot did not move for a while, or slightly moved in the direction opposite to the direction in which it should move.

This effect is particularly noticeable when the stage is moved in an open loop while looking at the sample observation image displayed on the image display device, and the operator feels uncomfortable with how the sample observation image moves, and There was an inconvenience that it was difficult to operate accurately. Further, conventionally, in order to rotate the observation visual field on the sample, the scanning direction of the electron beam is tilted. However, when the stage is mechanically moved to observe the area adjacent to the observation field of view in the state where the image on the image display device is rotated in this way, the reference of the moving direction of the stage is the coordinate before rotation. Since it is a system, there is a disadvantage that the sample observation image moves in the diagonal direction and is difficult to observe. For example, even if the observation field of view is rotated 45 degrees clockwise from the X axis and the observation field of view is moved in the direction of the left and right sides, the stage moves in the X direction. The image was moving in a direction tilted 45 ° counterclockwise from the X axis.

Further, in the conventional scanning electron microscope, when the observation visual field is moved by the mechanical visual field moving method for driving the stage while observing the sample observation image on the image display device, the observation image of the sample observation image is changed according to the observation magnification. Since the moving speeds are different, there is a discomfort in moving the sample observation image, and at the same time, it is difficult to position the observation visual field at an arbitrary position. Especially when the observation magnification is high,
Since the sample observation image moves rapidly, it was very difficult to position the observation visual field at an arbitrary position.

Further, in the conventional scanning electron microscope, since there is a pitch error of the feed screw (for example, ball screw) when the stage is moved, the observation field of view can be accurately positioned at a desired position by the open loop method. There was also the inconvenience that it was difficult. This occurs even if the pitch error of the feed screw is within the manufacturer's standard value, but the positioning accuracy required for the stage is stricter than the standard value. In particular, when the observation magnification is high, the range of the effective observation visual field corresponding to the effective screen of the image display device becomes extremely narrow, the positioning accuracy of the stage becomes extremely fine, and the pitch error of the feed screw cannot be ignored.

In view of the above point, the present invention is a scanning method that can accurately move an observation visual field on a sample to a desired position by a mechanical visual field moving method even if there is a backlash on a stage for positioning the sample. A first object is to provide a scanning electron microscope. Furthermore, in the present invention, when the observation visual field on the sample is electrically rotated, the observation visual field is moved in the direction of the adjacent region on the sample by the mechanical visual field movement method while observing the image display device. A second object of the present invention is to provide a scanning electron microscope in which the observation field of view can be moved without depending on the rotation angle of, and the observation field of view can be easily positioned.

Further, according to the present invention, when the observation visual field is moved by the mechanical visual field movement method based on the sample observation image of the image display device, the observation magnification of the sample does not depend on the magnification. The third object is to provide a scanning electron microscope capable of moving the sample observation image on the image display device at a constant speed. Further, according to the present invention, even if there is a pitch error in the feed screw of the stage for positioning the sample, the scanning visual field that can accurately move the observation visual field on the sample to the desired position by the mechanical visual field moving method. A fourth object is to provide an electron microscope.

[0012]

Means for Solving the Problems The first to fourth aspects of the present invention
As shown in FIG. 1, for example, the scanning electron microscope of (1) scans a sample (9) with an electron beam and detects the secondary electrons generated from the sample, and an image display means ( 25) In a scanning electron microscope that displays an image within a predetermined observation field on the sample (9), a feed screw stage that moves the sample (9) two-dimensionally on the scanning surface of the electron beam. (12X, 12Y, 13X, 13Y) and two pulse motors (14X, 14) that rotationally drive the two feed screws (13X, 13Y) of these stages, respectively.
Y) and microstep drive control means (20X, 20) for microstep driving these two pulse motors.
Y) and are common.

The first scanning electron microscope has a backlash amount storage means (33a) for storing the backlash amount when the moving direction of the stage is reversed, and its observation field of view on the sample (9). When moving by a predetermined movement amount, the predetermined movement amount is corrected based on the movement direction immediately before the stage, the movement direction from now on, and the stored contents of the backlash amount storage means (33a), and this correction is performed. The field-of-view movement control means (1) for driving the pulse motors (14X, 14Y) via the micro-step drive control means (20X, 20Y) by an angle corresponding to the moved amount.
9) and.

The second scanning electron microscope includes image rotating means (26 to 28) for rotating an image displayed on the image display means (25) by a predetermined angle, and its observation field of view on the sample (9). When moving the stage (12X, 12Y, 13X, 1) in the direction rotated by the image rotating means.
3Y) so that the microstep drive control means (2
0X, 20Y) through pulse motor (14X, 14Y
Y), and a visual field movement control means (19).

In the third scanning electron microscope, the magnification switching means (26) for switching the observation magnification of the image displayed on the image display means (25) and the observation field of view on the sample (9) are moved. In this case, the magnification switching means (2) is used for the moving speed before switching by the magnification switching means (26).
Microstep drive control means (20X, 2) so that the stages move at a speed obtained by multiplying the ratio of the observation magnification before switching by 6) and the future observation magnification.
And a visual field movement control means (19) for driving the pulse motors (14X, 14Y) via 0Y).

The fourth scanning electron microscope has a pitch error storage means (33d) for storing the pitch error between the two feed screws (13X, 13Y) of those stages, and its observation field of view on the sample (9). Is moved by a predetermined movement amount, the predetermined movement amount is corrected based on the pitch error stored in the pitch error storage means (33d), and the microstep is performed by an angle corresponding to the corrected movement amount. Field-of-view movement control means (1) for driving the pulse motors (14X, 14Y) via the drive control means (20X, 20Y).
9) and.

[0017]

According to the first to fourth scanning electron microscopes of the present invention, the stage for positioning the sample (9) has a micromotor driven pulse motor (14).
X, 14Y) is driven by a feed screw method. In the micro-step drive control, the minimum step angle of the pulse motor can be set to, for example, about 1/100 as compared with the full-step or half-step drive. Therefore, the positioning resolution is 1 or less than the width of the observation visual field set by the scanning electron microscope. It can be reduced by several digits.

First, in the first scanning electron microscope, the feed screw (13X, 12X, 12Y) of the stage (12X, 12Y) is used.
13Y) is obtained in advance and stored in the backlash amount storage means (33a), and the amount to be moved is corrected when the moving direction is reversed when driving the stages (12X, 12Y). . At this time, since the positioning resolution of the stage is high due to the micro-step drive, even if the backlash amount is about 1 μm, it can be accurately corrected.

Next, according to the second scanning electron microscope,
For example, as shown in FIG. 8A, when the observation field of view (43A) is electrically rotated on the sample by the angle θ with respect to the X axis, the screen (25a) of the image display means (25) displays the image shown in FIG.
As shown in (b), a magnified image of the rotated observation field (43A) is displayed. Then, when an instruction is issued to move the image in FIG. 8B in the right side direction, the stage (12
X, 12Y) moves in a direction inclined by an angle θ with respect to the X axis. As a result, the observation visual field (43A) gradually moves to the region (43B), the region (43C), .... At this time, the stages (12X, 12Y) that move by the micro step drive method compared to the normal width of the observation field (43A).
Since the resolution of can be finer by about one digit or more, the observation visual field can be moved accurately and smoothly.

According to the third scanning electron microscope,
Image display means (25) even when the observation magnification is switched
The moving speed of the sample observation image on the screen (25a) is constant. Therefore, the operability is improved. At this time, the stages (12X, 12Y) can be moved with high resolution by the micro-step drive method, so that the moving speed can also be controlled with high accuracy.

According to the fourth scanning electron microscope,
Stage (12X, 12Y) lead screw (13X, 13)
Y) pitch error is obtained in advance and pitch error storage means (3
3d), and the stage (12X, 12Y)
Correct the amount to move when driving. At this time, since the positioning resolution of the stage is high due to the micro step drive, even if the pitch error is about 1 μm, it can be accurately corrected.

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the scanning electron microscope according to the present invention will be described below with reference to the drawings. FIG. 1 shows a mechanical part and a control system of the scanning electron microscope of this embodiment. In FIG. 1, an electron beam 3 emitted from an upper electron gun 2 passes through an aperture 4 and a converging lens 5. After being converged, the X-deflection coil 6 and the Y-deflection coil 7 deflect the light in the X and Y directions, respectively. Note that, here, two directions orthogonal to each other on a plane perpendicular to the optical axis of the electron optical system are defined as an X direction and a Y direction.

The electron beam 3 deflected in two directions is further focused on the surface of a sample 9 in a sample chamber (not shown) via an objective lens 8 to scan a predetermined area on the surface. The sample 9 is placed on the Y stage 12Y via the sample table 11, and the Y stage 12Y is screwed onto the feed screw 13Y extending in the Y direction on the X stage 12X to form a pulse motor (that is, a stepping motor). 14Y to rotate the feed screw 13Y in the forward and reverse directions to move the Y stage 12Y to the Y direction.
Can be driven in any direction. Furthermore, the X stage 12X is a base 1
5, the X stage 12X can be driven in the X direction by being screwed into the feed screw 13X extending in the X direction and rotating the feed screw 13X in the forward and reverse directions by the pulse motor 14X. In this example, movement control devices 20X and 20Y for microstep driving drive pulse motors 14X and 14Y, respectively. That is, the pulse motors 14X and 14Y are each driven by a micro step drive method (details will be described later).

Further, a movable mirror 16X having a reflection surface perpendicular to the X axis and a movable mirror 16Y having a reflection surface perpendicular to the Y axis are fixed on the Y stage 12Y, and a laser interferometer 17X is provided.
Also, the Y coordinate of the sample stage 11 is measured by the movable mirror 16X, and the X coordinate of the sample stage 11 is measured by the laser interferometer 17Y and the movable mirror 16Y. The coordinates (X, Y) measured by the laser interferometers 17X and 17Y are supplied to the stage control means 19 in the central control system 18 that controls the operation of the entire apparatus.

In this embodiment, the stages 14X and 14Y are driven by a moving step close to the measurement resolution of the laser interferometer by the microstep driving method, so that it is not always necessary to provide the laser interferometers 17X and 17Y.
Therefore, in the present embodiment, the measurement values of the laser interferometers 17X and 17Y are monitored mainly when the pulse motors 14X and 14Y are out of step and the moving speeds and positions of the stages 14X and 14Y deviate from the designated states. I am using it for.

A secondary electron detector 22 is disposed diagonally above the sample 9, and the secondary electron 21 generated from the surface of the sample 9 by the irradiation of the electron beam 3 is detected by the secondary electron detector 22. It Detection signal S from the secondary electron detector 22
1 is stored in the frame memory in the image processing circuit 24 as image data through the amplifier 23 and an analog / digital converter (not shown). The image processing circuit 24 is provided with information indicating an observation visual field from an electron optical system control circuit 28, which will be described later.
And a synchronization signal corresponding to the scanning of the electron beam is also supplied,
The image processing circuit 24 reads out image data corresponding to the observation visual field from the frame memory and supplies it to the CRT display 25 together with a synchronization signal indicating scanning. On the display screen of the CRT display 25, the image in the observation visual field on the sample 9 is displayed at a predetermined magnification. Also, CR
On the display screen of the T display 25, a display for designating the enlargement magnification, the moving direction of the observation visual field, and the like is also provided.

Now, in order to use the scanning electron microscope,
It is necessary for the operator to set the enlargement magnification, the observation field of view, and the like. Therefore, in the central control system 18 of this example, the image magnification rotation visual field setting unit 26 for setting the enlargement magnification, the rotation angle of the observation visual field, and the observation visual field, and the stages 12X and 12Y.
Stage speed setting unit 29 for setting the moving speed and the moving direction of the stage, the stage position setting unit 30 for setting the moving direction and the moving distance of the stages 12X and 12Y, and the backlash of the stages 12X and 12Y as described later. A data input unit 31 for inputting a quantity and the like is connected.

Specifically, the image magnification rotation visual field setting unit 26
Is composed of a keyboard and a mouse, and the operator sets the numerical value of the observation magnification on the CRT display 25 by key input from the keyboard, sets the rotation angle of the observation visual field by key input, and displays it on the display screen of the CRT display 25. A desired portion is cut out from the enlarged image of the displayed sample by operating the mouse to set the observation visual field. The signals from the keyboard and mouse are supplied to the magnification setting storage means 27 in the central control system 18. The magnification setting storage means 27 stores the designated enlargement ratio, rotation angle, and position information of the contour of the observation field of view in the internal memory, and also the information of the enlargement ratio, rotation angle, and observation field of view is controlled by an electron optical system control circuit. 28.

From the electron optical system control circuit 28, a control signal corresponding to the designated enlargement magnification and the rotation angle of the observation field of view is sent to the electron optical system composed of the X deflection coil 6, the Y deflection coil 7 and the like. Is output. Further, the electron optical system control circuit 28 outputs to the image processing circuit 24 a synchronization signal synchronized with a control signal for scanning the electron beam by the X deflection coil 6 and the Y deflection coil 7, and information on the observation field of view. The electron beam scans the scanning range on the sample 9 according to the magnification, and rotates the scanning direction of the electron beam at a designated rotation angle. As a result, an enlarged image on the CRT display 25 is displayed at the magnification specified by the operator.
And the image of the sample in the state rotated by the designated rotation angle is displayed.

At this time, first, the CRT display 25
What is displayed on the display screen is an enlarged image of almost the entire scanning region of the sample 9 by the electron beam, and what is cut out by the mouse operation is an enlarged image of the partial region selected from the scanning region. is there. In this example, the area on the sample 9 corresponding to the magnified image displayed on the display screen of the CRT display 25 is called an “observation field of view”. Therefore, when a predetermined portion is cut out from the entire magnified image by operating the mouse, a variable partial area on the sample 9 corresponding to the cut out portion becomes the observation visual field. The image is cut out by reading out image data in a range corresponding to a specified observation visual field from the frame memory in the image processing circuit 24.

Further, the stage speed setting unit 29 is a so-called joystick and keyboard, and the operator operates the joystick as the stage speed setting unit 29 so that the moving direction of the stages 12X and 12Y,
And a signal indicating the moving speed is supplied to the stage control means 19. The stage control means 19 includes a movement control device 20.
The stages 12X and 12Y are driven at a designated speed via X and 20Y. Such a constant speed movement of the stage is performed, for example, when the operator wants to move the observation visual field in a desired direction while observing a magnified image of the sample on the CRT display 25. Further, the operator can directly input the moving direction and moving speed of the stage via the keyboard as the stage speed setting unit 29. An input device such as a so-called trackball may be used instead of the joystick.

On the other hand, the stage position setting section 30 is composed of a mouse and a keyboard, and, for example, the operator operates the mouse to specify a point (moving point) at a position away from the magnified image currently displayed on the CRT display 25. The information of the moving point and the current observation magnification and the position information of the observation visual field from the magnification setting storage means 27 are supplied to the stage control means 19, and the stage control means 19 moves from the center of the current observation visual field to the movement point. Calculate the direction and distance. Then, the stage control means 19 drives the stages 12X and 12Y by the calculated distance in the calculated direction via the movement control devices 20X and 20Y. As a result, the center of the observation visual field moves to the position designated by the operator.

Further, in the present example, the operator uses the keyboard as the stage position setting unit 30 to move, for example, 1 in the X direction.
When a command is input to move the stage by 0 mm, the stage control means 19 responds to the command by moving the pulse motor 14X through the movement control device 20X at a distance of 10 m.
It is designed to rotate by a rotation angle of m minutes.

Next, the data input section 31 comprises a keyboard, and the operator operates the stages 12X, 1 as described later.
Backlash amount of 3X and feed screws 13X, 13Y
After the pitch error is measured, the measured backlash amount and pitch error are input to the parameter setting means 32 in the central control system 18 via the data input unit 31. In response to this, the parameter setting means 32 stores the backlash amount and the pitch error in the backlash amount storage unit 33a and the pitch error storage unit 33d in the memory 33, respectively. After that, when driving the stages 12X and 12Y,
The stage control means 19 reads out the backlash amount and the pitch error from the memory 33 and corrects them. Other,
The stage control means 19 uses the previous stage 12X,
The traveling direction of 12Y, the observation magnification, the rotation angle of the observation visual field, and the like are written in the memory 33 via the parameter setting unit 32.

The magnification setting storage means 27, the stage control means 19, and the parameter setting means 32 in the central control system 18 in this example are functions executed by software of a computer. Here, the microstep drive control of the pulse motors 14X and 14Y in the present embodiment will be described in detail.

Generally, the pulse motor (stepping motor) rotates and stops at each step angle determined by the salient pole structure of the rotor and the stator, so that positioning control can be performed with higher accuracy and more easily than with a DC motor or an AC motor. It has the characteristic of being able to However, on the other hand, the pulse motor also has the characteristics that the rotor changes in speed at each step angle, so that the rotor changes speed and resonates or vibrates at a certain rotation speed.

On the other hand, the micro-step drive is subdivided by controlling the current flowing through the motor coil to control the basic step angle of the pulse motor to realize ultra-low speed drive, low vibration drive and low noise operation. This is a technology that can improve the positioning accuracy by increasing the positioning resolution. Specifically, in FIG. 2A, the basic step angle is 9 for two phases.
A pulse motor of 0 ° is shown, and in this FIG.
In a stator 51 made of a magnetic material having four convex portions formed at 90 ° intervals, a rotor 53 made of a magnet having opposite magnetic poles is arranged so as to rotatably face each other. The coil 52A is wound around one convex portion 51a of the stator 51, and the coil 52B is also wound around the convex portion 51b adjacent thereto.

When the pulse motor of FIG. 2 (a) is driven in full step, for example, first one coil 52
The current IA of A is set to the maximum positive value I ₀ , and the current IB of the other coil 52B is set to 0. As a result, the rotor 53 stands still at a position facing the convex portion 51a as indicated by the solid line. Next, when the current IA is set to 0 and the other current IB is set to the maximum positive value I ₀ , as shown by the dotted line, the rotor 53
Rotates 90 ° and stands still at a position facing the convex portion 51b. By alternately exciting the currents IA and IB in this manner, the rotor 53 rotates at a basic step angle of 90 °.

Next, when the pulse motor shown in FIG. 2A is driven by microstepping, the ratio between the current IA and the current IB is gradually changed. For example, as shown in FIG. 2B, the current IA is (1/4) I ₀ and the current IB is (3/4).
At I ₀ , the rotor 53 moves from the convex portion 51a to the convex portion 51b.
Stationary at a position rotated approximately (3/4) × 90 ° to the side.
Therefore, in principle, the minimum step angle can be made as small as possible with respect to the basic step angle by finely setting the ratio of the current IA and the current IB. This is microstep drive.

Further, in this embodiment, the pulse motor 14
As X and 14Y, as an example, 5 phases and a basic step angle φ
Uses a pulse motor of 0.72 °. Then, the basic step angle is further increased to 1 by the micro step drive.
Divide into about 00 divisions. A configuration example of the movement control devices 20X and 20Y for microstep driving used in this case will be described.

FIG. 3 shows the pulse motor 14X for the X-axis.
The phase control coils 52A to 52E and the movement control device 20X for controlling the currents flowing through these coils are shown. In FIG. 3, the movement control device 20X has the same configuration for independently supplying currents to the coils 52A to 52E. Current control circuit 5
4A to 54E are connected. For example, in the A-phase current control circuit 54A, the emitters of pnp transistors 56A and 58A are connected to the power supply terminal to which the DC voltage V _CC is applied, and the coil 52A is connected between the collectors of the transistors 56A and 58A. Together with npn transistors 57A and 5 in their collectors, respectively.
The 9A collector is connected. The emitters of the transistors 57A and 59A are commonly connected to one end of the resistor 60A, and the other end of the resistor 60A is grounded.
And Transistors 56A, 57A, 58A, 59A
And one end of the resistor 60A are connected to the A-phase control switching circuit 55A, and the A-phase control switching circuit 55A is connected to the transistors 56A, 57A, 58A,
The current flowing through each of 59A is adjusted.

Although not shown, the other coils 52B to 52B
The current control circuits 54B to 54E for E also include the same control switching circuit as the A-phase control switching circuit 55A. Also, the stage control means 19 of FIG.
The control signal SX from the phase-based switching circuit 61 of FIG.
The switching control circuit 61 supplies the current control circuits 54A to 54A so that the pulse motor 14X rotates by the rotation angle corresponding to the distance designated by the control signal SX.
The switching of each control switching circuit (55A etc.) in E and the current control operation are controlled. As a result, the pulse motor 14X has a minimum step angle of 0.72 ° / 10.
Rotate in the specified direction by the rotation angle specified by 0.
Further, the movement control device 2 for the Y-axis pulse motor 14Y
0Y has the same configuration as the movement control device 20X for the X axis.

As a movement control device for the pulse motors 14X and 14Y, there is a configuration example shown in FIG. 4 in addition to the configuration example shown in FIG. FIG. 4 shows another configuration example of the five-phase coils 52A to 52E of the pulse motor 14X and the movement control device 20X for controlling the currents flowing through these coils. In FIG. 4, the coils 52A to 52E are shown. Are serially connected to form a closed loop. The emitters of the pnp transistors 56A to 56E are commonly connected to the power supply terminal to which the DC voltage V _CC is applied, and the collectors of the transistors 56A to 56E are connected to one ends of the coils 52A to 52E, respectively. Npn transistor 57A to
The 57E collector is connected. The emitters of the transistors 57A to 57E are commonly connected to one end of the resistor 62, and the other end of the resistor 62 is grounded. And, Transistors 56A to 56E and 57A to 57E
Is connected to the control switching circuit 63, and the control switching circuit 63 regulates the current flowing through each of the transistors 56A to 56E and 57A to 57E.

In this case, the control signal SX from the stage control means 19 of FIG. 1 is supplied to the control switching circuit 63 of FIG. 4, and the control switching circuit 63 controls the control signal S.
The current flowing through each of the coils 52A to 52E is controlled so that the pulse motor 14X rotates by the rotation angle corresponding to the distance designated by X. As a result, the pulse motor 14X rotates in the designated direction by the designated rotation angle at the minimum step angle of 0.72 ° / 100.

Next, returning to FIG. 1, in this example, the feed screw 1
3X and 13Y are ball screws with 5 mm leads (5 mm pitch). In addition, as described above, the pulse motor 1
4X and 14Y each have 5 phases and basic step angle φ
Of 0.72 ° is used, and the basic step angle φ is divided into 100 divisions and driven by microstep driving. Therefore, the resolution p of the feed of the stages 12X and 12Y is 0.1 μm as follows.

P = 5 / (100 · 360 / 0.72) = 1 × 10 ⁻⁴ [mm] (1) If the observation magnification of the sample 9 is, for example, 100,000, the scanning area of the electron beam and its The width of the observation visual field on the sample 9 is 1.8 μm, for example, in relation to the observation magnification. On the other hand, since the resolution p of the feed of the stages 12X and 12Y is 0.1 μm, the observation visual field can be moved smoothly.

On the other hand, when the pulse motors 14X and 14Y are half-step driven by 4-5 phase excitation as in the prior art, the minimum step angle is 0.36 °, and therefore the resolution of the stages 12X and 12Y is advanced. 5 μm. Therefore, this conventional method cannot smoothly move the observation visual field having a width of 1.8 μm. Next, the operation of correcting the backlash amount of the stages 12X and 12Y in this embodiment will be described with reference to the flowchart of FIG. First, in step 101 of FIG. 5, a reference wafer is placed on the sample table 11 of FIG. This reference wafer is prepared for confirming the amount of movement of the stage, and scales (or marks) serving as references for the positions in the X direction and the Y direction are previously provided at predetermined intervals.

For example, in order to measure the amount of backlash in the X direction, in step 102, the operator instructs the stage position setting unit 30 to the stage control means 19 to move the X stage 12X from the current position by 10 mm (to be exact). It issues a command to move in the + X direction by 10.0000 mm, and so on. In response to this, the pulse motor 1
The 4X rotates by 10 mm by the microstep drive method. Thereafter, in step 103, the operator stores the image position of an appropriate scale on the reference wafer displayed on the CRT display 25. In this case, a predetermined index mark is displayed on the display screen of the CRT display 25, and the distance from the index mark to the image of the selected scale is recorded, for example.

Next, in step 104, the operator operates the X stage 12X via the stage position setting unit 30.
After moving 10 mm again in the + X direction using the open loop method,
In step 105, the X stage 12X is moved in the −X direction by the open loop method 10 via the stage position setting unit 30.
move mm. In subsequent step 106, the operator measures, on the display screen of the CRT display 25, how much the image position of the scale of the reference wafer stored previously and the image position of the current scale are deviated. At this time, + X from the index mark at the image position of the current scale
It is obtained by dividing the length (actual length on the display screen) obtained by subtracting the shift amount in the + X direction from the index mark obtained in step 103 from the shift amount in the direction by the observation magnification. The value is the backlash amount BX in the -X direction.
(-).

Since the observation magnification is 100,000 times, the backlash amount BX (-) can be accurately obtained on the order of 0.1 μm, for example. The operator uses the data input unit 3
1 and the backlash amount BX (−) via the parameter setting means 32 is stored in the backlash amount storage unit 33a in the memory 33. Similarly, the X stage 12
The backlash amount BX (+) when X is moved in the −X direction and then in the + X direction is also measured and stored in the backlash amount storage unit 33a. Furthermore, Y stage 1
Backlash amount BY when 2Y is reversed in the -Y direction
Backlash amount B when reversing in the (-) and + Y directions
Y (+) is also stored in the backlash amount storage unit 33a.

Next, using the calculated backlash amount,
The stage is controlled as follows. First, in step 107 of FIG. 5, the operator operates the stage position setting unit 3
For example, the moving direction (± X direction) of the X stage 12X and the moving distance ΔX are set from 0 to the stage control unit 19. At this time, the stage control means 19 previously stores information indicating the moving direction of the X stage 12X immediately before that in the backlash amount storage unit 33a. Thereafter, in step 108, the stage control means 19 checks whether or not the movement direction of the instructed X stage 12X has changed from the + X direction to the −X direction, and if such a movement direction has changed, in step 109, Backlash amount BX stored in the supplied movement distance ΔX
Moving distance (ΔX + BX) obtained by adding (-)
The pulse motor 14X is rotated by the open loop method by the amount of (-). As a result, the X stage 12X has its backlash amount corrected and accurately moves in the −X direction by ΔX.

When the movement direction changes from the -X direction to the + X direction in step 108, the operation proceeds to step 110 and the backlash amount BX (+).
The moving distance ΔX is corrected based on On the other hand, when the moving direction does not change, the moving distance ΔX is used as it is. For example, assuming that the backlash amount is 1 μm, by driving the pulse motor 14X of the present example by the microstep driving method, the positioning resolution of the X stage 12X is about 0.1 μm, so the backlash amount is accurately corrected. To be done.

On the other hand, in the conventional stage, even if the backlash amount was measured, it was difficult to correct the backlash amount. For example, even if the stage is provided with a ball screw with a 5 mm lead and the ball screw is driven in half step by a 5-phase pulse motor with 4-5 phase excitation, the positioning resolution is 5 μm per step. Therefore, if the backlash amount is 1
If it is about μm, it can be said that it is impossible to correct by the half-step driving method.

Next, an example of the operation of rotating the observation field on the sample 9 in this embodiment will be described with reference to FIGS. First, in FIG. 1, the sample 9 is placed on the sample table 11, and the observed image is displayed on the CRT display 25. FIG. 6A shows an observation visual field 4 on the sample 9 at that time.
2 shows the pattern 41 on the sample 9 in FIG.
A rectangular observation field 42 is set above. In addition, FIG.
(B) is a display screen 25 of the CRT display 25 of FIG.
a, and an enlarged image of the pattern in the observation visual field 42 is displayed on the display screen 25a. Therefore, when the magnification is β, the observation field of view 42 in FIG. 6A is actually 1 / β of the display screen 25a in FIG. 6B.

Next, the operator makes an angle θ (here, θ = 45 °) of the observation visual field 42 on the sample counterclockwise from the X direction via the image magnification rotary visual field setting unit 26 of FIG. Just rotate. As a result, the magnification setting storage means 27
The scanning direction of the electron beam on the sample 9 is 45
Rotate by 20 °, and as shown in FIG. 7A, a rectangular observation visual field 43 is set parallel to the direction tilted 45 ° counterclockwise with respect to the X direction on the pattern 41 of the sample 9. Further, the angle θ is stored in the rotation angle storage unit 33b in the memory 33 via the stage control unit 19. Then, an enlarged image of the pattern in the observation visual field 43 is displayed on the display screen 25a as shown in FIG. 7B. The enlarged image is displayed such that the sides of the rectangle are parallel or perpendicular to the sides of the display screen 25a.

Next, in FIG. 7B, the display screen 25
When observing the area on the right side of the enlarged image of a, the operator moves the stages 12X and 12Y on which the sample 9 is placed at a predetermined speed in the -X direction via the stage speed setting unit 29, for example. Issue a command to. In response to this, the stage control means 19 reads the angle θ of the rotation angle storage unit 33b of the memory 33, and the stages 12X and 12X.
Y is clockwise with respect to + X direction (180 ° -θ)
It is moved at a predetermined speed in the direction of rotation. This includes
For example, when the X stage 12X moves in the −X direction by ΔX, the Y stage 12Y may be moved in the −Y direction by ΔX · tan θ.

As a result, as shown in FIG. 8A, the observation field of view on the pattern 41 of the sample 9 is in the region 4 along the direction 44 inclined by the angle θ in the counterclockwise direction with respect to the + X direction.
3A, 43B, 43C, and so on. When the observation field of view moves to those areas 43A, 43B, 43C, CRT
The display screen 25a of the display 25 is shown in FIG.
Enlarged images as shown in (b), (c) and (d) are displayed. Therefore, according to this example, even when the observation field of view is electrically rotated, the observation field of view is set in the direction in which the observation field of view is set based on the enlarged image on the display screen 25a by the mechanical drive method without depending on the rotation amount. You can move.

At this time, the pulse motors 14X and 14Y are driven by the micro-step drive system, and the stage 12
Since the X and 12Y feed resolution is, for example, about 0.1 μm, the observation field of view can be moved smoothly even if the width of the observation field of view is 1.8 μm. Here, the prior art and the present embodiment will be compared. In the conventional technique, for example, the magnified image of FIG.
When trying to move in the X direction, the X stage 12X moved in the -X direction regardless of the rotation angle of the observation visual field. Therefore, the observation visual field moved in the diagonal direction.

FIG. 9A shows a state in which the rotated observation field of view is moved using the conventional technique. In FIG. 9A, the observation field of view on the pattern 41 of the sample 9 is along the + X direction. Area 45A, 45B, 45C, ...
When the observation field of view moves to those areas 45A, 45B, and 45C, enlarged images as shown in FIGS. 9B, 9C, and 9D are displayed on the display screen 25a of the CRT display 25, respectively. That is, the observation visual field moves in a direction unintended by the operator. This is because the operations of the stages 12X and 12Y do not follow the rotation information when the observation visual field on the sample is electrically rotated.

Further, in the conventional stage, since the sample observation image is moved without depending on the rotation amount of the visual field, even if the stage is moved by correcting the angle by the rotation angle of the observation visual field, the observation magnification is high. At magnification, it was difficult to move the image smoothly. For example, the feed screws 13X, 13
Pulse motor 1 with Y as a ball screw with a 5 mm lead
Even if 4X and 14Y are driven in half step by 4-5 phase excitation, the positioning resolution is 5 μm per step.
On the other hand, since the width of the observation visual field is 1.8 μm, the observation visual field cannot be moved smoothly.

Next, an example of the operation in the case of moving the observation visual field after switching the observation magnification in this example will be described. First, the operator sets the stage speed setting unit 29.
The speed of the sample observation image moving within the effective visual field range on the display screen of the CRT display 25 is determined via.
For example, the speed is set so that the sample observation image moves within the effective visual field range at time T (here, T is, for example, 3 seconds). Assuming that the observation visual field on the sample 9 is the observation visual field 42 having the width DX in the X direction shown in FIG.
Is moved in the direction of the area 47 adjacent in the + X direction at time T, the moving speed of the X stage 12X is −X.
The speed is set to DX / T in the direction. The observation magnification at this time is β ₀ (β ₀ is 10,000 times, for example). This observation magnification β
₀ is stored in the magnification storage unit 33c in the memory 33.

Next, when the operator changes the observation magnification to β (β is, for example, 20,000 times) via the image magnification rotation visual field setting unit 26, the changed observation magnification β is stored in the magnification storage unit 33c in the memory 33. Remembered. The magnification storage unit 33c also stores the previous observation magnification β ₀ . After that, when the operator indicates only the moving direction of the observation visual field from the keyboard as the stage speed setting unit 29, the stage control means 1
9 is a stage 12X, 12 by a micro step drive method so that the moving speed of the image on the display screen of the CRT display 25 becomes the moving speed at the observation magnification until then.
Move Y.

Specifically, in FIG. 10, the observation field of view 42
Is moved in the + X direction, the stage control means 19 drives the X stage 12X in the −X direction at a speed V (β) determined by the following equation. V (β) = (DX / T) β ₀ / β (2) As a result, the sample observation image moves at the time T within the effective visual field range on the display screen of the CRT display 25. Therefore, in this example, the moving speed of the magnified image of the sample on the display screen becomes constant without depending on the observation magnification, and the operability is improved.

On the other hand, in the prior art, when the moving speed of the stage is first set, the moving speed thereafter is maintained at the initial set value. Therefore, for example, if the sample observation image is initially set to move within the effective visual field range of 10,000 times on the display screen in 3 seconds, the time required to move the effective visual field range at the observation magnification of 20,000 times. Becomes 1.5 seconds, and the time becomes 0.3 seconds at an observation magnification of 100,000 times. This is not convenient for moving the sample observation image.

Furthermore, in the prior art, it was difficult to move the observed image of the sample at a constant speed. For example, CR
Move within the effective field of view of the T display 25 in 3 seconds,
The observation magnification at this time is 10,000 times. At this time, as the feed screw 13X, 13Y 5mm lead ball screw,
If the pulse motors 14X and 14Y are driven in half step by 4-5 phase excitation, the positioning resolution of the stages 12X and 12Y will be 5 μm per step. Also, 1
If the width DX of the observation visual field on the sample 9 at a magnification of 10,000 is set to 18 μm due to the relationship between the scanning width of the electron beam and the observation magnification, the number of steps that can be moved within the effective visual field range is three.
The speed of the stage is set to move 3 steps in 3 seconds. Next, when the observation magnification is 100,000 times, the width of the observation visual field becomes 1.8 μm. In this case, since the positioning resolution is 5 μm per step, if the stage is moved by one step, a completely different observation image will be displayed on the CRT display 25. Therefore, it is difficult to move within the effective visual field range in 3 seconds when the observation magnification is 100,000 times.

On the other hand, when the micro-step driving is used as in this embodiment, the positioning resolution of the stages 12X and 12Y is 0.1 μm, so that the observation magnification is 10.
The observation field of view can be moved at a constant speed without any problem even if the magnification is 10,000 times. Next, in this embodiment, the feed screw 13
An example of the operation when correcting the X, 13Y pitch error will be described. First, the reference wafer is placed on the sample table 11 in FIG. In order to confirm the movement amount of the stage, the reference wafer is preliminarily formed with a scale serving as a reference for the movement distance. The scale is calibrated in advance by a highly accurate laser interferometer or the like. This reference wafer and the stage 12X via the pulse motors 14X and 14Y,
By comparing the movement amount of 12Y, the feed screw 13
It is possible to measure the pitch error amounts of X and 13Y.

For example, the X stage 12X has an X coordinate of 0.
Open loop control from 0mm position only 5.0mm +
Suppose you want to move in the X direction. At this time, the X coordinate is 0.0 m
The reference scale of 0.0 mm is set in the vicinity of the index mark at the center of the display screen of the CRT display 25 at m, and the amount of misalignment between the index mark and the reference scale is read. Next, the pulse motor 14X is driven by the micro-step driving method to move the X stage 12X in the + X direction by 5.0 mm, and the 5.0 mm reference scale of the reference wafer is displayed on the CRT display 25. Therefore, the difference between the deviation amount of the reference mark from the index mark at the center of the screen and the deviation amount of the previously read reference scale of 0.0 mm is divided by the observation magnification to obtain from 5. The pitch error ΔP (X) of the feed screw 13X up to 0 mm can be obtained. By performing this operation for every 5.0 mm in the X direction, the pitch error ΔP (X) of the feed screw 13X in the X direction can be measured.

FIG. 11 shows an example of the pitch error ΔP (X) measured in this way. In FIG.
The pitch error ΔP (X) changes along the coordinate X in a polygonal line 48 shape. Similarly, the pitch error ΔP (Y) of the feed screw 13Y in the Y direction can be measured. These pitch errors ΔP (X) and ΔP (Y) are stored in the pitch error amount storage unit 33d in the memory 33 of FIG.

The stage is controlled as follows by using the obtained pitch error amount. First, the operator sets the moving direction and moving distance of the stages 12X and 12Y to the stage control means 19 via the stage position setting unit 30. The stage control means 19 determines the X stage 12X and the Y stage 12Y from the moving direction and the moving distance.
After calculating the amount of movement of, the amount of movement is corrected using the pitch errors ΔP (X) and ΔP (Y) in the pitch error storage unit 33d.

For example, referring to FIG. 11, the X stage 12
When X moves from the position X ₁ to the position X ₂ , the stage control means 19 calculates a difference (ΔP (X ₂ ) −ΔP (X) between pitch errors in the calculated moving amount (X ₂ −X ₁ ). ₁ )) is added. Then, the stage control means 19 drives the pulse motor 14X by the microstep drive method and the open loop method by the corrected movement amount. As a result, the X stage 12X accurately moves by the distance (X ₂ −X ₁ ), so that it is not affected by the pitch error of the feed screw 13X.
The observation field of view can be moved to the target value on the sample. On the other hand, in the conventional stage, even if the pitch error amount of the feed screw was measured, it was difficult to correct this pitch error amount. . This is because the pitch error amount is usually smaller than the positioning resolution of the conventional stage, and therefore the pitch error cannot be corrected by the open loop method.

However, if the micro-step drive system is used as in this embodiment, the feed screws 13X and 13Y are provisionally used.
Even if the pitch error is about 1 μm, the positioning error of the stage is about 0.1 μm, so that the pitch error can be accurately corrected. In the above embodiment, the measurement values of the coordinates (X, Y) by the laser interferometers 17X and 17Y are used for detecting the step-out of the pulse motors 14X and 14Y. However, a simpler detector (such as an optical linear encoder) may be used for detecting out-of-step. Furthermore, if the operation of the pulse motor is stable, it is not necessary to provide a laser interferometer or the like.

As described above, the present invention is not limited to the above-described embodiments, and various configurations can be adopted without departing from the gist of the present invention.

[0073]

According to the present invention, since the sample is positioned on the stage driven by the feed screw system by the pulse motor controlled by the micro step drive system, the positioning resolution of the sample is the observation field of the scanning electron microscope. Since the width can be reduced by about one digit or more from the normal width, the following operation is possible.

That is, according to the first scanning electron microscope,
Even if there is backlash on the stage for positioning the sample, the observation field of view on the sample can be accurately moved to a desired position by the mechanical field of view moving method. Further, according to the second scanning electron microscope, the observation field of view is directed toward the adjacent region on the sample by the mechanical field of view moving method while observing the image display device while electrically rotating the observation field of view on the sample. When moving to, the observation visual field can be moved without depending on the rotation angle of the observation visual field, and the positioning of the observation visual field is easy.

According to the third scanning electron microscope,
When the observation field of view is moved by the mechanical field-of-view movement method based on the sample observation image of the image display device, no matter how many times the observation magnification of the sample is, it does not depend on the magnification and at a constant speed on the image display device. The sample observation image can be moved. Furthermore, according to the fourth scanning electron microscope, even if there is a pitch error in the feed screw of the stage for positioning the sample, the observation field of view on the sample can be accurately set to the desired position by the mechanical field of view moving method. It can be moved.

[Brief description of drawings]

FIG. 1 is a configuration diagram including a partial perspective view showing an embodiment of a scanning electron microscope according to the present invention.

FIG. 2A is an explanatory diagram when driving a pulse motor by a full-step drive system, and FIG. 2B is an explanatory diagram when driving a pulse motor by a micro-step drive system.

FIG. 3 is a configuration diagram showing an example of wire connection of coils of a pulse motor 14X of FIG. 1 and a movement control device 20X for microstep drive control.

FIG. 4 is a configuration diagram showing another possible example of connection of coils of the pulse motor 14X of FIG. 1 and a movement control device 20X for microstep driving.

FIG. 5 is a flowchart showing an example of a backlash correction method in the embodiment.

FIG. 6 is a diagram showing an observation visual field on a sample and an enlarged image thereof.

FIG. 7 is a diagram showing an enlarged image observed when the observation visual field is rotated in the example.

FIG. 8 is an operation explanatory diagram when the rotated observation visual field is laterally moved in the example.

FIG. 9 is an operation explanatory diagram when the rotated observation field is laterally moved by the conventional technique.

FIG. 10 is an explanatory diagram when the observation field of view is moved at a constant speed on the display screen.

FIG. 11 is a diagram showing an example of the pitch error of the feed screw of the embodiment.

[Explanation of symbols]

2 Electron gun 4 Aperture 5 Converging lens 6 X deflection coil 7 Y deflection coil 8 Objective lens 9 Sample 11 Sample stage 12X X stage 12Y Y stage 13X, 13Y Feed screw 14X, 14Y Pulse motor 17X, 17Y Laser interferometer 18 Central control system 19 Stage Control Means 20X, 20Y Movement Control Device for Microstep Drive Control 22 Secondary Electron Detector 24 Image Processing Circuit 25 CRT Display 26 Image Magnification Rotation Field-of-View Setting Section 27 Magnification Setting Storage Means 29 Stage Speed Setting Section 30 Stage Position Setting Part 31 Data input part 42, 43, 43A, 43B Observation field of view

Claims (4)

[Claims] 1. An image in a predetermined observation visual field on the sample is displayed on an image display means by using an image signal obtained by scanning an electron beam on the sample and detecting secondary electrons generated from the sample. In a scanning electron microscope for displaying, a feed screw type stage for moving the sample two-dimensionally on the scanning surface of the electron beam and two feed screws of the stage are rotationally driven.
One pulse motor, a microstep drive control means for microstep driving the two pulse motors, a backlash amount storage means for storing the backlash amount when the moving direction of the stage is reversed, and the above-mentioned sample on the sample. When the observation visual field is moved by a predetermined movement amount, the predetermined movement amount is corrected based on the movement immediately before the stage, the movement direction from now on, and the stored contents of the backlash amount storage means, and the correction is performed. And a field-of-view movement control unit that drives the pulse motor through the micro-step drive control unit by an angle corresponding to the amount of movement, and a scanning electron microscope. 2. An image in a predetermined observation visual field on the sample is displayed on the image display means by using an image signal obtained by scanning an electron beam on the sample and detecting secondary electrons generated from the sample. In a scanning electron microscope for displaying, a feed screw type stage for moving the sample two-dimensionally on the scanning surface of the electron beam and two feed screws of the stage are rotationally driven.
One pulse motor, a micro-step drive control means for micro-step driving the two pulse motors, an image rotation means for rotating an image displayed on the image display means by a predetermined angle, and the observation field of view on the sample. And a field-of-view movement control means for driving the pulse motor via the micro-step drive control means so that the stage advances in the direction rotated by the image rotation means when the scanning is performed. Electron microscope. 3. An image in a predetermined observation visual field on the sample is displayed on the image display means by using an image signal obtained by scanning an electron beam on the sample and detecting secondary electrons generated from the sample. In a scanning electron microscope for displaying, a feed screw type stage for moving the sample two-dimensionally on the scanning surface of the electron beam and two feed screws of the stage are rotationally driven.
One pulse motor, a microstep drive control means for microstep driving the two pulse motors, a magnification switching means for switching an observation magnification of an image displayed on the image display means, and an observation visual field on the sample. When moving, with respect to the moving speed before being switched by the magnification switching means,
A visual field movement that drives the pulse motor through the micro-step drive control means so that the stage advances at a speed obtained by multiplying the ratio of the observation magnification before switching by the magnification switching means and the future observation magnification. A scanning electron microscope comprising: a control unit. 4. An image in a predetermined observation visual field on the sample is displayed on the image display means by using an image signal obtained by scanning an electron beam on the sample and detecting secondary electrons generated from the sample. In a scanning electron microscope for displaying, a feed screw type stage for moving the sample two-dimensionally on the scanning surface of the electron beam and two feed screws of the stage are rotationally driven.
One pulse motor, a microstep drive control means for microstep driving the two pulse motors, a pitch error storage means for storing a pitch error between the two feed screws of the stage, and a predetermined observation field of view on the sample. Of the above-mentioned amount of movement, the predetermined amount of movement is corrected based on the pitch error stored in the pitch error storage means, and the microstep drive control means is moved by an angle corresponding to the corrected movement amount. And a field-of-view movement control unit that drives the pulse motor via the scanning electron microscope.