JP2020177924A - Charged particle beam device - Google Patents

Abstract

To provide a charged particle beam device which allows for field of view movement to an accurate position, even when performing field of view movement on an actual specimen.SOLUTION: A charged particle beam device includes an electron source (1) generating a charged particle beam irradiating a specimen of measurement object, a field of view movement deflector (8) for deflecting the charged particle beam, a stage (12) for mounting the specimen, and a control arrangement acquiring the images of the specimen for multiple patterns thereof by moving the field of view by means of the deflector, or acquiring multiple images at respective positions of the patterns, and adjusting a signal supplied to the deflector so as to correct the rotation angle of the image, based on the results of pattern matching of the design data of the pattern and the multiple images thus acquired.SELECTED DRAWING: Figure 1

Description

The present disclosure relates to a charged particle beam device, and more particularly to a charged particle beam device including a deflector for moving a visual field position by deflecting a charged particle beam.

In a charged particle beam device such as a scanning electron microscope, a technique of moving the scanning position of an electron beam on a sample by a deflector (hereinafter, also referred to as image shift) is known. This technique can move the scanning position with higher accuracy than the technique of moving the scanning position of the electron beam by moving the stage. In Patent Document 1, an image shift is performed on a sample in which a plurality of patterns are arranged at predetermined known intervals, a difference between a reference position in an image and a center of gravity position of the pattern is obtained, and the difference is compensated for. A charged particle beam apparatus that accurately shifts an image by performing an image shift is disclosed.

Patent No. 5164355 (corresponding US Pat. No. USP7,935,925)

As described in Patent Document 1, the scanning position is positioned on a plurality of objects by using image shift, and the amount of deflection of the electron beam is corrected based on the deviation of the objects, whereby the visual field movement accuracy of the image shift is achieved. Can be improved. However, when the field of view is moved by image shifting on an actual sample (for example, a semiconductor wafer), the field of view may not be positioned at an appropriate position even if the above correction is performed.

Below, we propose a charged particle beam device that aims to enable the field of view to move to an accurate position even when the field of view is moved on an actual sample.

As one aspect for achieving the above object, a charged particle source for generating a charged particle beam to irradiate a sample to be measured, a deflector for deflecting the charged particle beam, and a stage for mounting the sample. Then, the field of view is moved by the deflector to acquire an image for each of the plurality of patterns of the sample, or a plurality of images are acquired at each position of the pattern, and the design data of the pattern and the plurality of acquired images are acquired. We propose a charged particle beam device including a control device that adjusts a signal supplied to the deflector so as to correct a rotation angle deviation of the image based on a result of pattern matching with an image.

According to the above configuration, even if the magnetic field for adjusting the charged particle beam fluctuates, it is possible to perform highly accurate visual field movement.

The figure explaining the outline of the scanning electron microscope provided with the field-of-view movement deflector. The figure explaining the positional relationship between a sample to be measured and a reference pattern, and an example of a reference pattern. The flowchart which shows the process of calculating the rotation angle deviation by the magnetic field of an electromagnetic lens, and correcting the amount of current and voltage value applied to the field of view moving deflector. The figure explaining the example of the image acquired to calculate the rotation angle deviation by the magnetic field of an electromagnetic lens. The figure explaining the outline of the scanning electron microscope provided with the field-of-view movement deflector. The figure explaining the outline of the scanning electron microscope provided with the field-of-view movement deflector.

In recent years, device design and manufacturing processes have become more complicated with the miniaturization and three-dimensional structure of semiconductor devices. In the start-up of a complicated manufacturing process, it is necessary to inspect and measure a large amount of fine patterns. A charged particle beam device such as a scanning electron microscope is used for inspection and measurement of such a semiconductor device. A scanning electron microscope is a device that performs measurement and inspection using an image or the like obtained by scanning a focused electron beam on a sample.

In order to perform many inspections and measurements in a short time, it is necessary to capture the inspection points and measurement points in the field of view (Field Of View: FOV) of the scanning electron microscope at high speed. For such visual field movement, an image shift technique capable of moving the scanning position of the electron beam at a higher speed and with higher accuracy than the stage movement is suitable. When performing image shift, prepare a sample in which patterns are arranged in known predetermined interval units, move the field of view using an image shift deflector so that these patterns are captured in the field of view, and calibrate the deflection signal. By doing so, it is possible to perform highly accurate image shift, but in reality, even if such calibration is performed, it may not be possible to properly move the field of view. The focus condition of the sample to be actually measured or inspected changes depending on the thickness of the sample and the charge attached to the sample. If the sample used for calibration of the deflection position and the sample to be measured or inspected are different, the focus condition may be different, the condition of the electromagnetic lens for focusing the beam will also change, and the change in the condition will be the deflection accuracy. Affects. In other words, when the control parameters (control current) set in the electromagnetic lens or deflector are changed, the generated magnetic field also changes compared to when the deflection position is calibrated by image shift, so high deflection accuracy is achieved. It may become unsustainable. In addition, it is conceivable that the influence of hysteresis and temperature change of the magnetic material used for the electromagnetic lens cannot be ignored.

In the examples described below, a charged particle beam device capable of highly accurate visual field movement will be described regardless of the influence of hysteresis, temperature change, etc. of the magnetic material used for the electromagnetic lens.

In the embodiment described below, for example, an objective lens that focuses a charged particle beam emitted from a charged particle source and irradiates a sample, a field moving deflector that deflects the charged particle beam, and a measurement target. A charged particle beam device provided with a reference pattern in a region where the visual field can be moved by moving the stage from the sample, and the focus is adjusted by changing the control parameter (control current) of the electromagnetic lens on the sample to be measured. After moving to the reference pattern by the stage while maintaining the control parameters, a plurality of images of the reference pattern are acquired while changing the visual field movement amount of the image shift, and the rotation angle due to the magnetic field of the electromagnetic lens is calculated from the acquired images. After correcting the current amount or voltage value applied to the field moving deflector from the rotation angle, the charged particle beam device moves to the measurement point on the sample by the stage and starts the inspection and measurement of the measurement target. explain.

According to such a configuration, even if the magnetic field generated by the charged particle beam device changes between the sample to be measured or inspected and the sample for calibration, the field of view is accurately moved by image shifting. It becomes possible.

Hereinafter, a charged particle beam device (charged particle beam device) provided with a field view moving deflector will be described with reference to the drawings. FIG. 1 shows a schematic view of a scanning electron microscope equipped with a field-of-view moving deflector. The scanning electron microscope illustrated in FIG. 1 is provided with a field-of-view movement deflector including an electromagnetic deflector 8 and an electrostatic deflector 9. These field-of-view movement deflectors do not necessarily have to have both electromagnetic and electrostatic types, and may have only one of them. Further, both the electromagnetic type and the electrostatic type may be provided with two or more deflectors. The primary electrons 51 drawn from the electron source 1 by the first anode 2 are accelerated by the second anode 3, focused by the first capacitor lens 4, and then pass through the objective diaphragm 5. After that, it is focused by the second condenser lens 6 and then focused on the sample 11 by the objective lens 10.

As a method of moving the field of view between different measurement points, a method of positioning the focusing point of the primary electron 51 at a desired position on the sample 11 by driving the sample stage 12 for arranging the sample 11 and an electromagnetic deflector 8 or A method (image shift) of moving the field of view (scanning region) of a scanning electron microscope by electrically deflecting the primary electron 51 by the electrostatic deflector 9 and changing the arrival position of the primary electron 51 in the sample 11 is there. Since the stage movement involves mechanical movement, it is difficult to move the field of view at high speed. Furthermore, the visual field movement accuracy is lower than that of image shift. However, a high-quality image can be obtained even if a large visual field movement of 100 mm or more is performed. On the other hand, image shift has the advantage that the field of view can be moved accurately in a shorter time than the stage movement because the amount of deflection is controlled by the current and voltage applied to the deflector. For example, when the field of view is moved by several tens of μm or more. Since the beam is largely deviated from the ideal optical axis 50, the image quality may deteriorate due to the off-axis aberration of the objective lens 10.

When measuring a sample with a high density of measurement points such as hotspot analysis of a logic device, if both image shift of a large area of several hundred μm and high-quality image acquisition can be achieved, multiple measurement points can be set within the range of the image shift. You can catch it. Therefore, the number of stage movements during measurement is reduced, and the time required for multipoint measurement can be significantly shortened. Further, with the miniaturization of semiconductor devices, there is an increasing demand for measurement and inspection of fine patterns of 10 nm or less. In order to support inspection and measurement of such fine patterns, it is required to improve the measurement accuracy by acquiring an image at a high magnification. However, when an image is acquired at a high magnification, the field of view becomes narrow, and it becomes difficult to capture a fine pattern in the field of view. Therefore, it is desired to move the field of view accurately by using image shift rather than stage movement.

However, even if the image shift is used, it is very difficult to maintain the visual field movement accuracy of 10 nm or less while performing the image shift in a large region of several hundred μm. This is because scanning electron microscopes often use an electromagnetic lens as an objective lens in order to increase the resolution, and the visual field movement accuracy of image shift is greatly affected by the magnetic field fluctuation of this electromagnetic lens. For example, when performing an image shift of 100 μm, in order to maintain the visual field movement accuracy of 10 nm or less, the rotation angle of the primary electron 51 due to the magnetic field must be precisely adjusted with an accuracy of 0.005 °. It is difficult to avoid the fluctuation of the magnetic field of the electromagnetic lens because it is generated by the hysteresis and the temperature change of the magnetic material constituting the electromagnetic lens. Therefore, even if the visual field movement amount is adjusted with high accuracy in advance, if the relationship between the control parameter when observing the sample to be measured and the magnetic field actually generated is slightly different from that at the time of adjustment, the visual field of the image shift. The movement accuracy will be reduced.

Hereinafter, as the first embodiment, even when the magnetic field changes due to the hysteresis or temperature change of the magnetic material contained in the electromagnetic lens on the sample to be measured, the visual field is accurately moved by image shifting. An image acquisition sequence or a processing sequence for measurement and inspection will be described. The scanning electron microscope as illustrated in FIG. 1 is controlled by a control device (not shown). The control device includes an image calculation processing device and a storage medium for storing an operation program for automatically executing a processing sequence described later. The image processing device generates an image based on the detection signal acquired by the scanning electron microscope according to the operation program, and executes necessary arithmetic processing. In addition, the control device executes control of a mechanical system such as a stage and control of an optical system such as beam deflection and focusing according to an operation program. This also applies to the examples described later.

FIG. 2 is a diagram showing an outline of a stage 12 on which a sample 11 to be measured or inspected is placed. A space for mounting the reference pattern (standard sample or calibration sample) 13 is provided on the stage 12. In this way, both the sample 11 and the reference pattern 13 are installed on the stage 12, and by moving the stage, both can be positioned directly below the ideal optical axis 50. An example of the reference pattern 13 is shown in FIG. 2 (b).

P001, 002, 003, 004 indicate a sample in which a specific pattern is periodically arranged. By calibrating the visual field movement position using the patterns arranged in the two-dimensional direction in this way, it is possible to perform calibration regardless of the visual field movement direction. Further, P005, 006, 007, 008 exemplify a reference pattern including a line pattern long in a specific direction. With the recent miniaturization of patterns, the line width of line patterns tends to become narrower. That is, the ratio of the length of the line pattern to the line width tends to be larger. In order to measure the entire line pattern at high magnification (narrow field of view), it is necessary to appropriately move the high-power field of view along the line pattern. Calibration using patterns as illustrated in P005, 006, 007, 008 makes it possible to move the visual field with high accuracy when measuring a long pattern in a specific direction over a wide range. It is also possible to use a radial pattern as shown in P009, 010, 011 and 012. Here, when using an inclined pattern such as P007, 008, 010, 011 and 012, it is desirable to know the inclined angle in advance.

The measurement processing procedure using the calibration sample as described above will be described with reference to the flowchart of FIG. As described above, a storage medium (memory) (not shown) stores an operation program that automatically executes the process illustrated in FIG. 3, and the control device is a mechanical system such as a stage according to this operation program. It controls the electron microscope optical system and the image processing device.

First, the stage is moved so that the measurement point 30 on the sample 11 to be measured is located directly below the ideal optical axis 50 (S001). The measurement point 30 may be a place where there is a pattern to be actually inspected and measured, or the center position of the sample 11 or the like may be selected and used as a representative position. At this time, it is desirable that the amount of visual field movement due to the deflection of the electron beam (the amount of visual field movement due to image shift) be as small as possible. In addition, the location where the focus condition is the same as or close to the pattern to be actually measured (the location where the sample height and charging conditions are the same or approximate).

Next, the desired optical conditions used for the measurement are set (S002). Here, the optical conditions indicate the amount of current and the voltage value applied to the electron source 1, the first anode 2, the second anode 3, the first capacitor lens 4, the second capacitor lens 6, and the objective lens 10. Next, the focus is adjusted by changing the current value supplied to the objective lens 10 which is an electromagnetic lens (S003). At this time, it is desirable to select the optimum magnification for focus adjustment according to the pattern shape of the measurement point 30. Next, the stage is moved so that the field of view of the electron beam is positioned at the position of the reference pattern 13 (S004). At this time, care should be taken not to change the amount of current applied to the objective lens 10 set in S003. That is, an image of the reference pattern is generated while maintaining the focus condition adjusted by the focus evaluation using the image obtained by irradiating the sample to be measured (measurement target pattern) with the electron beam. This is because the visual field movement condition may change by changing the focus condition (magnetic field condition).

If focus adjustment is required when moving to the position of the reference pattern 13, it is performed by changing the control value of one of the electrostatic lenses. Examples of the control value of the electrostatic lens include a voltage value applied to the objective lens 10 and a voltage value applied to the sample 11. Next, an image of the reference pattern 13 is acquired for the visual field movement amount IS due to a plurality of different image shifts (S005).

FIG. 4 shows an example of the image to be acquired. Here, the case of a long line pattern will be described. First, the image 40 is acquired at IS = (X1, Y1). Here, X1 and Y1 indicate the amount of visual field movement in the X and Y directions of IS, respectively, and the values of X1 and Y1 when acquiring the image 40 are arbitrary. When IS = (X, Y) is input, a control device (not shown) calculates the current supplied to the visual field moving deflectors 8 and 9 and the voltage value to be applied. An example of the calculation formula used here is shown below (Equation 1).

Here, Vx and Vy indicate the amount of current and the voltage value applied to the field view moving deflectors 8 and 9. In addition, values obtained by experiments or calculations are set in advance for each of the parameters a, b, c, and d.

Next, the IS is input so that the field of view moves along the line pattern. In the case of a line pattern that is long in the vertical direction (Y direction), IS = (X1, Y2) is input (here, Y1 ≠ Y2), and the image 41 is acquired. Hereinafter, the image 42 and the like are acquired while changing the IS in the same manner. IS = (X1, Y3) when acquiring the image 42 is arbitrary, but it is desirable that the difference between Y1 and Y3 is large. However, in order to more accurately identify the visual field fluctuation according to each image shift position, it is desirable to increase the image acquisition amount of different visual field movement amounts.

Next, from the images 40 and 42 acquired in S005, the image deviation Δr in the direction perpendicular to the line pattern is calculated (S006). Here, Δr is the x direction (line) of the image 42 between the image center (xc, yc) of the image 42 and the line pattern center 70 when the central axis of the line pattern is located at the center of the image in the image 40. The distance is in the direction of minute 63). In specifying the center position of the pattern using the image 40, if the center of the image and the center axis of the line pattern do not match, the difference value between the center position of the pattern in the image 40 and the pattern center coordinates and the image center. Obtain Δrb, include the difference value, and perform an operation as described later. That is, Δr (Δrb) is calculated for the image 40 in the same manner as described above, and the difference from Δr in the image 43 is used as Δr by image shift in the following calculation.

Next, the rotation angle deviation Δθ is calculated based on the Δr calculated in S006 (S007). Since Δθ and Δr have a relationship of ΔIS × Δθ = Δr, the rotation angle deviation Δθ can be calculated by Δθ = Δr / ΔIS. Here, ΔIS is the difference in the amount of visual field movement due to the image shift input when the images 40 and 42 are acquired, and in the example of S005, it is the absolute value of the difference between Y1 and Y3. Next, based on the calculated rotation angle deviation Δθ, at least one of the current value supplied to the visual field moving deflectors 8 and 9 and the applied voltage value is corrected by using the following equation (Equation 2) (S008). ..

Further, in S004, when the focus adjustment is performed by changing the control value of the electrostatic lens by ΔV, the correction is performed using the following equation (Equation 3).

θΔV is a change in the rotation angle due to a magnetic field by changing the control value of the electrostatic lens by ΔV, and MΔV is a change in the magnification of the electrostatic lens by changing the control value of the electrostatic lens by ΔV. Here, each parameter of A and B is set with a value obtained in advance by experiment or calculation. Here, the control formula of the image shift is corrected using the parameters (A, B) acquired in advance, but by performing the focus adjustment when moving to the reference pattern 13 with the electrostatic lens, the influence of hysteresis and the like is performed. It is possible to make corrections without receiving.

Next, in S009 and 010, by performing image shift, image acquisition (S009), and rotation deviation calculation (S010) on the reference pattern by the deflection signal corrected by the correction signal obtained as described above. It is confirmed whether the rotation angle deviation Δθ satisfies the specification value (S011). That is, verification (confirmation) is performed as to whether or not the correction using the reference pattern is properly performed.

Here, the specification value Δθth is Δθth = ISp / ISmax, where ISp is the visual field movement accuracy of the desired image shift, and ISmax is the maximum image shift visual field movement amount used when inspecting or measuring the sample to be measured. It is desirable to obtain from. ISmax may calculate the amount of visual field movement due to the maximum image shift expected from the inspection points and the positional relationship of the measurement points in the sample to be measured, the order in which the inspections and measurements are performed, and the like. Degradation of resolution due to off-axis aberration of the objective lens that occurs when image shift is operated, maximum output voltage of power supply that applies voltage and current to deflectors 8 and 9 for field of view movement, current value, output limitation by control software, etc. Numerical values may be diverted.

If the rotation angle deviation Δθ calculated in S010 is smaller than the specification value Δθth, the vehicle moves to the measurement point 30 by moving the stage, and inspection and measurement of the sample to be measured are started. If the rotation angle deviation Δθ is larger than the specification value Δθth, the process returns to S008 and the process is repeated until the specification value is satisfied.

By using the above procedure, even if the relationship between the current value applied to the electromagnetic lens and the magnetic field actually generated changes due to the hysteresis or temperature change of the electromagnetic lens, it can be accurately measured on the sample to be measured. It is possible to move the field of view by image shifting.

The second practical example relates to a method of accurately moving the field of view by image shifting even when the sample to be measured is charged by irradiation with an electron beam. When the sample is charged, the focusing point of the electron beam shifts from the sample surface, making it difficult to inspect and measure with high image quality, so focus adjustment is required. By this focus adjustment, the rotation angle due to the magnetic field changes. The change in the rotation angle is any of the current value applied to the objective lens 10 for focus adjustment, the voltage value applied to the objective lens 10, and the negative voltage 14 (hereinafter, also referred to as retarding voltage) applied to the sample 11. The same occurs even if is used.

The reason why the rotation angle changes even if an electrostatic lens is used is that the energy of electrons when the magnetic field passes through a certain region is changed. Hereinafter, a charged particle beam apparatus capable of performing correction processing when the sample is charged will be described using the configuration shown in FIG.

The charged particle beam device illustrated in FIG. 5 is equipped with a surface electrometer 15 (Surface Potential Measurement: SPM) for measuring the surface potential of the sample. The potential of the sample can be measured by using SPM. In addition, the arithmetic processing unit provided in the control device can store the potential at each position on the sample and obtain the average potential of a predetermined region of the sample.

Hereinafter, a specific sequence will be described based on the flow shown in FIG. For S001 and 002, the same procedure as in Example 1 is performed. In S003, after adjusting the focus by changing the amount of current of the objective lens 10, the potential VW of the sample to be measured is measured using SPM. Here, instead of using SPM, the potential measurement is performed based on the image obtained when the negative voltage 14 applied to the sample is set lower than the energy of the electron beam and the electron beam is driven back before reaching the sample. You may. As another means, an energy filter is installed between the reflector 7 and the sample 11, or in the case of a detector that directly detects the electrons emitted from the sample, between the detector and the sample. By calculating the difference between the curve (S-curve) showing the relationship between the voltage when the retarding voltage is swept and the detected amount of electrons and the S-curve showing the state without charge acquired in advance, charging is performed. You can also find the amount.

For S004, 005, 006, and 007, the same procedure as in Example 1 is performed. Next, when correcting the deflection signal in S008, when an electrostatic lens is selected for focus adjustment when the stage is moved to the position of the reference pattern 13 in S004, the correction formula differs depending on the type of the electrostatic lens. .. As an example, a case where the voltage value applied to the objective lens 10 is selected and a case where the negative voltage 14 applied to the sample is selected will be described. First, a case where the focus adjustment is performed by changing the voltage value applied to the electrostatic lens included in the objective lens by ΔV will be described. The rotation angle deviation Δθ obtained in S007 is affected by the voltage change ΔV at the time of focus adjustment and the charge VW of the sample. In order to accurately move the field of view by image shifting in the sample 11 to be measured, it is necessary to use a correction formula that takes these effects into consideration. The correction formula (Equation 4) in this case is shown below.

Here, each parameter of A1, A2, B1 and B2 is set with a value obtained in advance by experiment or calculation. It should also be noted that A1 and A2 are different values. Similarly, B1 and B2 have different values.

Next, a case where the focus adjustment is performed by changing the negative voltage 14 applied to the sample by ΔV will be described. In this case, since both the rotation angle deviation due to charging and the rotation angle deviation due to focus adjustment are due to changes in the sample potential, the correction formula (Equation 5) is as follows.

Here, each parameter of A1 and B2 is set with a value obtained in advance by experiment or calculation. It should be noted that a common coefficient is used when calculating θAV and MAV from the charge VW of the sample to be measured and the change amount ΔV of the negative voltage 14 for focus adjustment.

Hereinafter, S009, 010, 011 and 012 are carried out in the same manner as in Example 1. By using such control, even if the potentials of the sample 11 to be measured and the reference pattern 13 are different, the deviation of the rotation angle due to the magnetic field of the objective lens can be precisely controlled, and the image shift can be performed accurately. It is possible to move the field of view by.

In the third embodiment, the sample to be measured is inspected, the decrease in the visual field movement accuracy of the image shift is detected during the measurement, and the image shift input amount and the current amount applied to the visual field movement deflector. It relates to a sequence for readjusting the relationship between voltage values. In the inspection and measurement of a semiconductor device, a large amount of inspection and measurement may be required for the same sample, and it may take several tens of hours or more from the start of measurement to the end of measurement. In such a long-time inspection and measurement, the sample height may change due to expansion and contraction of the stage 12 due to changes in the device environment (atmospheric pressure, air temperature, etc.), and the visual field movement accuracy of the image shift may decrease. According to this embodiment, it is possible to accurately move the field of view by image shift even in the inspection and measurement for a long time.

In the inspection and measurement of semiconductor devices, the part to be observed is determined in advance, and it is often possible to perform pattern matching between the pattern shape design data and the acquired image. In this embodiment, when the field of view is moved between these patterns by image shift, the acquired image and the design data are pattern-matched. Pattern matching may be performed at all observation points, or may be performed by designating an interval such as one point for several points. Alternatively, conditions such as when a certain time has passed since the previous pattern matching may be set. When such a condition is set, it is possible to prevent deterioration of the visual field movement accuracy due to the image shift even when the pattern matching between the acquired image and the design data cannot be performed.

If the field of view movement due to image shift is accurate, the pattern to be measured can be captured in the center of the field of view of the SEM, but if the field of view movement accuracy decreases, the pattern will be located at the edge of the FOV or the pattern cannot be captured in the FOV. Can be considered. In such a case, the sequence shown in FIG. 3 is executed (that is, the rotation angle of the image shift at the time of visual field movement using the reference pattern is determined), and the image shift input amount and the visual field movement deflector are applied. Readjust the relationship between current and voltage values. In other words, it is determined whether or not the visual field should be moved to the reference pattern according to the result of pattern matching. It should be noted here that the visual field movement of the image shift may be accurate even if the pattern is not located within the FOV. For example, when the pattern is not formed, or when the pattern is formed but the position is different from the design. Even in such a case, the pattern may not be formed as designed by acquiring the image of the same observation point again after executing the sequence shown in FIG. 3, or the visual field movement accuracy of the image shift. It is possible to isolate whether the image has deteriorated, and in the former case, it is possible to feed back to the device manufacturing process.

By using such a sequence, it is possible to detect the deterioration of the visual field movement accuracy due to the image shift and readjust it. Therefore, even in the inspection and measurement for a very long time, the visual field movement accuracy of the high image shift can be achieved. Can be maintained. Further, the control method of this embodiment is an image not only when the temperature and atmospheric pressure change, but also when the charged state changes temporally or spatially, or when the pattern shape is deformed by the irradiation of the charged particle beam. It can be used as a means for readjusting the amount of shift visual field movement.

As a trigger for readjustment, a matching failure (a matching score pattern having a predetermined value or more did not exist in the visual field, a plurality of matching score patterns having a predetermined value or more existed in the visual field, etc.) can be considered. However, the difference between the center of the visual field (or the original matching position) and the matching position may be monitored, and readjustment may be performed when the difference exceeds a predetermined value. In addition, the timekeeping change of the matching position is monitored, a function showing the relationship between the matching position information and the time is created, the time to reach a predetermined deviation amount is predicted by extrapolation to the function, and the readjustment time is set. You may set it.

The fourth embodiment relates to stabilizing the secondary electron detection efficiency when the visual field is moved by image shifting. First, a method for detecting secondary electrons in a scanning electron microscope will be described with reference to FIG. The secondary electrons 52 emitted from the sample collide with the reflector 7 having a hole in the center, and new secondary electrons 53 are emitted from the arrival position of the secondary electrons 52 on the reflector 7. After that, the secondary electrons 53 are drawn by a negative voltage (not shown) applied to the detector 17, then converted into an electric signal and detected.

Further, the secondary electrons 54 deflected by the secondary electron deflector 16 may collide with the reflector 7', and similarly, the signal electrons emitted from the reflector 7'may be detected by the detector 17'. As an example of the backscatter for secondary electrons 16, there is a Wien filter that deflects only the secondary electrons without deflecting the primary electrons by superimposing the deflection fields of the electric field and the magnetic field. When the secondary electrons 52 are detected by using the detector 17', the secondary electrons 52 are directly deflected and detected by the secondary electron deflector 16 without using the reflectors 7 and 7'. You can also do it.

In such a secondary electron detection mechanism, in order to discriminate the emission angle and energy of the secondary electrons 52, the secondary electron deflector 16 is appropriately controlled and emitted from the sample 11 at a desired emission angle and energy. It is important that the secondary electrons 52 are stably passed through the holes formed in the reflector 7. When the field of view is moved by image shifting, the amount of current and the voltage value applied to the secondary electron deflector 16 are changed according to the amount of field of view movement, so that the secondary electrons of the desired emission angle and energy are always used. Can be detected stably.

However, if the relationship between the control value of the objective lens 10 and the magnetic field actually generated changes, the rotation angle of the secondary electrons 52 also changes, so that the secondary electrons of the desired emission angle and energy are selectively detected. I can't. In this embodiment, a method of appropriately controlling the orbits of secondary electrons with a desired emission angle and energy by the secondary electron deflector 16 will be described even when the rotation angle of the magnetic field changes due to the hysteresis or temperature change of the electromagnetic lens. To do.

The amount of current or voltage applied to the secondary electron deflector 16 is controlled according to the amount of visual field movement IS = (X, Y) due to image shift and the amount of current (or its control value) supplied to the objective lens 10. .. Therefore, when the relationship between the amount of current supplied to the objective lens 10 (or its control value) and the magnetic field actually generated changes, the amount of current or voltage applied to the secondary electron deflector 16 is also optimized accordingly. There is a need to.

For this purpose, the rotation angle deviation Δθ obtained when the sequence shown in FIG. 3 is performed is used. This Δθ is a relative angle (angle deviation) between the direction of the image obtained by experiments or calculations in advance and the image obtained when a predetermined current is actually supplied to the objective lens 10. If there is no discrepancy between the relationship between the amount of current and the magnetic field acquired in advance and the relationship between the actual amount of current and the magnetic field, the rotation angle deviation Δθ obtained by the sequence of FIG. 3 is 0. Therefore, it is possible to correct the relationship between the amount of current applied to the objective lens 10 (or its control value) and the magnetic field actually generated from the rotation angle deviation Δθ. The correction formula for the amount of current or the voltage value (ALX, ALY) applied to the secondary electron deflector 16 is as follows (Equation 6).

Here, the values of α, β, γ, and ζ are set in advance by experiments or calculations. Further, CX and CY are offset values for passing secondary electrons of a desired emission angle and energy through the holes of the reflector 7 in the field of view movement amount 0 due to the image shift. Further, CSE is a coefficient that associates the rotation angle deviation Δθ of the primary electrons 51 with the rotation angle deviation ΔθSE of the secondary electrons. The primary electrons 51 and the secondary electrons 52 have different energies when passing through the electromagnetic field of the objective lens 10. Therefore, the rotation angle deviation Δθ obtained by the sequence of FIG. 3 cannot be used as it is for correcting the current amount or voltage value applied to the secondary electron deflector 16.

Here, the coefficient CSE needs to be changed when the amount of current applied to the electromagnetic lens and the electrostatic lens and the voltage value change. The CSE may be obtained experimentally or calculated. Further, it is necessary to use different values for each parameter used here when the optical conditions (current amount applied to each lens, voltage value, etc.) are changed.

By using the above control, secondary electrons emitted from the sample at a desired emission angle and energy can be selectively selected even when the orbits of secondary electrons change due to hysteresis or temperature change of the electromagnetic lens. It becomes possible to detect.

1: Electron source, 2: First anode, 3: Second anode, 4: First condenser lens, 5: Objective aperture, 6: Second condenser lens, 7, 7': Reflector, 8: Electromagnetic field movement Deflector, 9: Electrostatic deflector for moving the visual field, 10: Objective lens, 11: Sample, 12: Stage, 13: Reference pattern, 14: Returning voltage application part, 15: Surface electrometer, 16: Secondary electrons For deflectors, 17, 17': detector, 30: measurement point, 40, 41, 42: field of view of acquired image, 50: optical axis, 51: primary electrons, 52, 52, 53: secondary electrons, 60, 61, 62, 63: Central axis of image

Claims (13)

A charged particle source that generates a charged particle beam to irradiate the sample to be measured,
A deflector that deflects the charged particle beam and
A stage for placing the sample and
The field of view is moved by the deflector to acquire an image for each of the plurality of patterns of the sample, or a plurality of images are acquired at each position of the pattern, and the design data of the pattern and the plurality of acquired images are used. A charged particle beam device including a control device that adjusts a signal supplied to the deflector so as to correct a rotation angle deviation of the image based on the result of pattern matching. In claim 1,
The control device is a charged particle beam device, characterized in that a second pattern matching is performed after a certain period of time has elapsed from the first pattern matching. In claim 1,
The control device is a charged particle beam device that adjusts a signal supplied to the deflector according to a comparison result between a matching score indicating the result of the pattern matching and a threshold value. In claim 1,
The control device is a charged particle beam device, which sets a time for adjusting a signal supplied to the deflector based on a function indicating a change over time as a result of the pattern matching. In claim 1,
The control device is a charged particle beam device, characterized in that the pattern matching is performed at set time intervals, and the visual field is moved to the pattern based on the result of the pattern matching. In claim 1,
The deflector is a charged particle beam device, which is a field-of-view movement deflector. In claim 6,
The control device calculates the field of view movement accuracy when the field of view is moved by the field of view movement deflector based on the result of the pattern matching, and the field of view moves according to the comparison result between the field of view movement accuracy and the threshold value. A charged particle beam device characterized by adjusting the signal supplied to a mobile deflector. In claim 6,
The control device calculates the visual field movement accuracy when the visual field is moved by the visual field movement deflector based on the result of the pattern matching, and moves the visual field according to the comparison result between the visual field movement accuracy and the threshold value. A charged particle beam device for adjusting the relationship between the amount and the amount of current and the voltage value applied to the field-of-view movement deflector. In claim 1,
The control device is a charged particle beam device that adjusts a signal supplied to the deflector based on the result of the pattern matching during execution of a sequence including a step of acquiring the plurality of images. In claim 9.
The control device is a charged particle beam device characterized in that pattern matching is performed at one or more observation points specified in the sequence. In claim 9.
The control device is characterized in that, when it is determined from the result of the pattern matching that the pattern is not located in the field of view at a specific observation point, an image at the specific observation point is acquired after executing the sequence. Charged particle beam device. A charged particle source that generates a charged particle beam to irradiate the sample to be measured,
A deflector that deflects the charged particle beam and
A stage for placing the sample and
The field of view is moved by the deflector to acquire an image for each of the plurality of patterns of the sample, or a plurality of images are acquired at each position of the pattern, and the design data of the pattern and the plurality of acquired images are used. A charged particle beam device including a control device for determining the necessity of moving the field of view to the pattern based on the result of pattern matching. A charged particle source that generates a charged particle beam to irradiate the sample to be measured,
A deflector that deflects the charged particle beam and
A stage for placing the sample and
The field of view is moved by the deflector to acquire an image for each of the plurality of patterns of the sample, or a plurality of images are acquired at each position of the pattern, and the design data of the pattern and the plurality of acquired images are used. A charged particle beam device including a control device for moving a field of view to the pattern based on the result of pattern matching.