Device and method for aberration correction that works with a beam of charged particles

Abstract

A device operating with a beam of charged particles, which has the following features: a source (1) for a beam (2) of charged particles, an optical system operating with charged particles, designed to irradiate a sample (10) with charged particles emitted from the source (1) for a beam (2) of charged particles, an aberration corrector (6) designed to correct the aberration of the charged particle optical system, a control unit (26) designed to control the charged particle optical system and the aberration corrector (6), a through-the-focal-point imaging unit (18) designed to obtain, in a first step, several Ronchigrams in which the focal point position of the charged particle optical system is changed, an aberration calculation unit designed to subdivide the obtained Ronchigram into several local areas in a second step and to calculate the aberration magnitude based on focal lines detected in the local areas in a third step, and an input unit (15) designed for inputting a setting value, characterized by the fact that the through-focal-point imaging unit (18) is designed to change the focal point position within a range based on the entered setting value, The aberration calculation unit is designed to detect the focal lines in the local areas of each obtained Ronchigram on the side below the focal point and on the side above the focal point, and the through-focal-point imaging unit (18) is designed to obtain multiple Ronchigrams when the focal point position is changed in an area larger than the area based on the input setting value, when the focal lines detected in the local areas are smaller than a defined value.

Description

Technical field

[0001] The present invention relates to a device operating with a beam of charged particles and a method for aberration correction, and in particular to a device operating with a beam of charged particles which is provided with an aberration corrector. State of the art

[0002] In recent years, a device operating with a beam of charged particles, similar to a transmission electron microscope (TEM) or a scanning transmission electron microscope (STEM), has been proposed, which incorporates an aberration corrector. The aberration corrector is used to correct the spherical aberration (Cs) of an objective lens, which is a major factor limiting the resolution in a TEM or STEM. An aberration corrector is known that comprises two multipole lenses generating a hexapole magnetic field and two axially symmetric lenses (transmission lenses) arranged between them. Furthermore, the Cs of the objective lens is corrected by the aberration corrector.

[0003] An additional aberration, known as parasitic aberration, is generated by imperfections in the aberration corrector, namely a positional shift of each pole forming the multipole lens or a variation in a magnetic property of a pole material. The third-order or lower-order parasitic aberration generated in this way includes first-order bi-symmetric astigmatism (A1), second-order singly symmetric coma aberration (B2), second-order triple-symmetric astigmatism (A2), third-order bi-symmetric stellar aberration (S3), third-order quadruple-symmetric astigmatism (A3), or the like. When adjusting the aberration corrector, these parasitic aberrations must be corrected in addition to the correction of Cs.The aberration corrector is adjusted by repeating the aberration measurement to quantify the magnitude of any aberration remaining in the optical system and by applying an aberration correction based on the measurement result. A method for aberration measurement is disclosed, for example, in PTL 1. List of citations from patent literature

[0004] PTL 1: JP 2007 / 180 013 A Summary of the invention: Technical problem

[0005] PTL 1 discloses a technique in which the aberration remaining in an optical system is observed as a change in a local region of a Ronchigram. Furthermore, an observer changes a focal point near an exact focal position to obtain a Ronchigram in which there is no line with a directional direction in the central part, thereby correcting double astigmatism. That is, in aberration measurement, a characteristic aberration magnitude is extracted from several electron microscope images obtained by changing the focal point or the tilt angle of a beam, and aberration coefficients are calculated that specify the magnitude and direction of the aberration. However, if the aberration is very large because the electron microscope images are highly distorted, the characteristic magnitude is difficult to extract.For example, the first-order bi-symmetric astigmatism (hereinafter referred to as A1) and the second-order singly symmetric coma aberration (hereinafter referred to as B2) can be significantly altered by an operation such as switching the accelerating voltage or changing the sample. If, in this case, a user observes the change in the electron microscope image, as shown in PTL 1, while shaking the focal point and manually adjusting the deflection of the electron beam or the like, such that the characteristic patterns of A1 or B2 are reduced, the following problems arise: If large values ​​remain for A1 and B2 because the distortion resulting from A1 and B2 also becomes large, it is impossible to detect the change in the characteristic patterns without significantly altering the adjustment values ​​of A1 and B2. Consequently, a user inexperienced with adjustments finds themselves in a situation where they cannot detect the change in the patterns and therefore cannot easily adjust them correctly while appropriately determining the adjustment value for a specific aberration to be corrected. This is because, in a state where the change in the patterns is not detected, even when comparing images before and after adjustment, it cannot be determined whether the adjustment was correct before and after.One object of the invention is to provide a device operating with a beam of charged particles, whereby the aberration can be measured even in a state in which an aberration amount with large values ​​(for example A1 and B2) remains, which cannot be easily measured by the known technique.

[0006] A conventional aberration correction method is disclosed in the publication “Measurement method of aberration from Ronchigram by autocorrelation function”, Ultramicroscopy 108 (2008), pages 1467-1475 by H. Sawada et al. The two-part version of the present claims 1 and 6 is based on this publication.

[0007] Another conventional method is described in the publication “Aberration measurement using the Ronchigram contrast transfer function”, Ultramicroscopy 110 (2010), pages 891-898 by AR Lupini et al. Solution to the problem

[0008] A device operating with a beam of charged particles according to this application has the features specified in claim 1. Claim 6 protects the associated method. The dependent claims relate to preferred embodiments. Advantageous effects of the invention

[0009] According to the invention, even an aberration that is not easily measured using known techniques can be corrected. Brief description of the drawing

[0010] They show: Fig. 1 a configuration diagram of an example embodiment of the invention, Fig. 2. A diagram of the relationship between a Ronchigram and off-axis defocusing. Fig. 3. A diagram of the relationship between a local autocorrelation function of a Ronchigram and off-axis astigmatism. Fig. 4. A diagram showing the direction of off-axis astigmatism. the Fig. 5A and Fig. 5B Flowcharts of a procedure for the automatic correction of C1, A1 and B2, Fig. 6A a diagram of an example for calculating the total sum of the areas of the fitting ellipses of local autocorrelation functions, Fig. 6B a diagram of an example for calculating the total area of ​​the fitting ellipses of local autocorrelation functions, Fig. 7A a diagram of an example of focal line detection on the side below the focal point, Fig. 7B a diagram of an example of focal line detection on the side above the focal point, Fig. 8 A diagram illustrating a comparison of detection accuracies of focal lines due to the difference in focal point spacing widths, Fig. 9. A diagram illustrating an example of a procedure for taking central coordinates in a local area and Fig. 10. A diagram of an example GUI, which performs automatic aberration correction. Description of embodiments

[0011] According to embodiments of the invention, several Ronchigrams obtained by performing through-focusing are used. Each Ronchigram is subdivided into several local areas to identify a focal line for each local area, thereby calculating off-axis defocus C1(τ) and off-axis astigmatism A1(τ). Furthermore, an aberration coefficient is calculated using the least squares method based on C1(τ) and A1(τ), and the aberration is corrected based on the result. One embodiment of the invention is described below with reference to the accompanying drawing.

[0012] Fig. Figure 1 is a configuration diagram showing an example of the embodiment of the invention. Based on Fig. Figure 1 describes a scanning transmission electron microscope (STEM), a device that operates with a beam of charged particles. Irradiation lenses 3 and 4 act on electron beams 2 emitted by an electron beam source 1 to form a focal point on the object plane of a focusing lens 5. The electron beams 2 enter an aberration corrector 6 parallel to each other due to the action of the focusing lens 5. Any aberration of the electron beams 2 is corrected in the aberration corrector 6, and a focal point is formed above an objective lens 9 as a result of the action of a transfer lens 7. The electron beams 2, whose aberration is corrected, converge through the objective lens 9 to form a fine probe on a sample 10. Scanning coils 8 are used to scan the sample 10 using the probe.Scattered waves emanating from sample 10 are then received by a dark-field image detector 12. The intensity of the scattered waves is then displayed as a bright dot sequence in an input / output unit 15, synchronously with the probe scanning by a dark-field image observation unit 14, in order to obtain a dark-field image of sample 10.

[0013] When a Ronchigram is observed, the scanning by the electron beam using the scanning coils 8 is interrupted, the transmitted and scattered waves emerging from the sample 10 are received by a camera 13, and a two-dimensional distribution of the intensities of the transmitted and scattered waves is displayed as a bright dot sequence in the input and output unit 15 by the Ronchigram observation unit 16. A projection lens 11 adjusts the broadening of the electron beams entering the dark-field image detector 12 and the camera 13.

[0014] An automatic aberration correction device 17 comprises a through-the-focal-point imaging unit 18, an input image validity determination unit 19, a correction necessity determination unit 20, a unit 21 for calculating off-axis aberration, a unit 22 for calculating on-axis aberration, a correction control amount calculation unit 23, a conversion coefficient table 24, and an optimal focal point range calculation unit 25. The through-the-focal-point imaging unit 18 commands the Ronchigram observation unit 16 to import a Ronchigram as an input for the automatic aberration correction device 17, while the focal point is changed by modifying the excitation of the objective lens 9 by a control unit 26.

[0015] The input image validity determination unit 19 determines the validity of the Ronchigram image acquired using Ronchigrams with different focal points, in particular whether the Ronchigram image was acquired in an area where the sample is located. The correction necessity determination unit 20 determines, based on the degree of distortion of the Ronchigram, whether it is necessary to perform the correction according to the present embodiment in order to avoid unnecessary correction.

[0016] Unit 21, for calculating off-axis aberration, divides each of the multiple Ronchigrams obtained by transmission focusing into several local areas and detects a focal line for each local area in order to calculate the off-axis defocus C1(τ) and the off-axis astigmatism A1(τ). Unit 22, for calculating on-axis aberration, calculates an aberration coefficient using the least-squares method based on C1(τ) and A1(τ).

[0017] The correction tax amount calculation unit 23 calculates the tax amount for each lens to correct the aberration with respect to the conversion coefficient table 24, based on the aberration coefficients calculated by unit 22 for calculating on-axis aberration, and outputs the tax amount to the control unit 26. The optimal focal point range calculation unit 25 calculates the optimal through-the-focal-point range and the optimal focal point spacing width based on the aberration coefficients calculated by unit 22 for calculating on-axis aberration. It is not necessary for the focal point spacing width to be constant and uniform within the through-the-focal-point range. The interval of the focal point spacing width can be changed according to a focal line condition.

[0018] Here, the relationship between the Ronchigram and off-axis defocusing is discussed with reference to... Fig. As described in Figure 2, when considering an electron beam flux 1 passing through the vicinity of the axis and an electron beam flux 2 passing through an outer part of the objective lens, the electron beam converges on the front side compared to electron beam flux 1 because the electron beam passing through the outer part of the lens is strongly bent by the influence of spherical aberration Cs. The height of the focal point formed by electron beam flux 1, i.e., light rays on the axis, is the defocus C1 on the axis, and the height of the focal point formed by electron beam flux 2, i.e., light rays off the axis, is the defocus C1(τ) off the axis.

[0019] If two-dimensional coordinates of a point where the center (dashed line in) is located are used, Fig. 2) where the light beam intersects a projection plane (central point in a local region) off-axis, (u, v) are, then τ is u + iv (i is an imaginary unit). If the distance between the sample surface and the projection plane (detection area of ​​a camera) is L (L can be changed by exciting the projection lens 11), the magnification of a projection image of a sample produced in a local region 1 by the electron beam flux 1 is M1 = (C1 + L) / C1 and the magnification of a projection image of the sample produced in a local region 2 by the electron beam flux 2 is M2 = (C1(τ) + L) / C1(τ). Because the relationships L >> C1 and L >> C1(τ) hold, in general M1 ≅ L / C1 and M2 ≅ L / C1(τ).This means that the projection image of the sample observed in each local area has a magnification that is inversely proportional to any local defocusing (defocusing on-axis in an on-axis case and defocusing off-axis in an off-axis case).

[0020] Next, the relationship between an autocorrelation function in a local area of ​​a Ronchigram (hereinafter referred to as the local autocorrelation function) and the off-axis astigmatism will be discussed with reference to Fig. 3 described. Fig. Figure 3(a) shows isointensity lines of local autocorrelation functions when through-focusing is performed (when a focal point is changed in several states over a specific region). Here, lu and lv represent local coordinates, with a central point of a local region being the origin. The size of the ellipse formed by the isointensity line of the local autocorrelation function is proportional to the projection magnification of the local region.

[0021] As described above, the envelope of the isointensity line changes in a plane lu = 0 and a plane lv = 0 in a form proportional to the -1-th power of the focal point, because the projection magnification of the local area is inversely proportional to the defocusing. If the reciprocal of the isointensity line coordinates is taken here, as in Fig. As shown in Figure 3(b), the envelope of the isointensity line becomes a straight line. In the plane lu = 0 and in the plane lv = 0, the positions at which the envelopes form foci are different from each other, and the ellipses formed by the isointensity lines when the focus is formed in a plane become linear (hereinafter referred to as the focus line).

[0022] The focal line appears at two points on the side below the focus and on the side above the focus, with an exact focus point lying between them. This represents a state in which the focus point lies on an axis. An interval (astigmatism difference) of the foci at two locations represents the magnitude |A1(τ)| of the off-axis astigmatism. That is, |A1(τ)| can be expressed by formula (1) if the focal point magnitudes are denoted as f1 (side below the focus) and f2 (side above the focus) when a focal line is formed: |A1(τ)|=1 / 2(f2−f1)

[0023] The direction ∠A1(τ) of the off-axis astigmatism is reflected in the direction of the ellipses formed by the isointensity lines of the local autocorrelation functions. As in Fig. As shown in Figure 4, formula (2) is obtained if the angle formed by the long axis of the ellipse on the side above the focus and the lv-axis is called θ: ∠A1(τ)=θ

[0024] Furthermore, the deviation from the exact focal point in each local area before through-focusing (in the initial state), i.e., the defocusing C1(τ) off-axis, is given by formula (3): C1(τ)=−1 / 2(Δf1+Δf2)

[0025] According to the present embodiment, off-axis defocus and off-axis astigmatism are calculated with reference to the through-focused Ronchigram by fitting ellipses to the isointensity lines of the local autocorrelation functions and determining the focal point values ​​f1 and f2 at the focal line where the ellipse extends furthest, and the inclination θ of the long axis of the ellipse at f2. Off-axis defocus and off-axis astigmatism are expressed using the on-axis aberration coefficients by formulas (4) and (5). C1(τ)=C1 +2(B2τ*+B2*τ) +2C3|τ|2+3(S3τ*2+S3*τ2) +2(3B4|τ|2τ+B4*τ3)+12D4|τ|2τ*+4A4τ*3 +2C5|τ|2τ2+5A5τ*4+2(6S5|τ|4)+20R5|τ|2τ*2*2 A1(τ)=A1 +2B2τ+2A2τ* +C3τ2+6S3|τ|2+3A3τ*2 +2(3B3|τ|2τ+B4*τ3)+12D4|τ|2τ*+4A4τ*3 +2C5|τ|2τ2+5A5τ*4+2(6S5|τ|4)+20R5|τ|2τ*2*2

[0026] Here, fifth-order or lower-order aberrations are considered as on-axis aberration coefficients (B4: fourth-order bi-symmetric coma aberration, D4: fourth-order triple-symmetric Trefoil aberration, A4: fourth-order five-symmetric astigmatism, C5: fifth-order spherical aberration, A5: fifth-order six-symmetric astigmatism, S5: fifth-order bi-symmetric stellar aberration, R5: fifth-order quadruple-symmetric Rosetta aberration). By generating simultaneous equations relating to off-axis defocusing and off-axis astigmatism with respect to each local region, the on-axis aberration coefficient can be calculated using the least-squares method.

[0027] The following is an operating sequence of the automatic aberration correction device 17 with reference to the flowchart from the Fig. 5A and Fig. 5B is described. Here, C1, A1, and B2 are used as targets for automatic correction. First, a through-the-focal-point area T and a focal-point distance width Δf are defined (F1). T is set so that this value lies between the focal lines on the side below the focal point and the side above the focal point. According to this setting, the number of images captured through the focal point becomes T / Δf + 1.

[0028] Next, a Ronchigram 1 is obtained in an initial state (F2). Then, a Ronchigram 2 is obtained where the focal point is changed by -nΔf (n ≡ T / (2Δf)) (F3). Furthermore, a correction value is calculated for Ronchigram 1 and Ronchigram 2 (F4). If Ronchigrams 1 and 2 are obtained in an area where no sample is present, the correlation value will be large because the image contrast hardly changes.

[0029] Accordingly, the correlation value is compared with a specific threshold R (F5). If the correlation value is greater than R, the correction is stopped. If the correlation value is less than R, a local autocorrelation function of Ronchigram 1 is calculated (F6), and a total sum of the areas of fitting ellipses is calculated (F7).

[0030] Fig. Figure 6 shows examples of Ronchigrams, local autocorrelation functions, and the total area of ​​the fitting ellipses of the local autocorrelation functions. The upper diagram in Fig. Figure 6A shows an example where a obtained Ronchigram is subdivided into local regions. The lower diagram shows an example where the total area of ​​the fitting ellipses in each subdivided region is calculated. The upper and lower diagrams in Fig. 6B have the same relationship. Here, defining the local regions of the Ronchigram is related to the order of the aberration to be corrected. That is, a lower-order aberration appears in a central section of the Ronchigram, and a higher-order aberration appears in a peripheral section, which is separated from the central section of the Ronchigram. Accordingly, it is possible to set the type (the order) of an aberration that can be corrected according to the procedure for defining the local regions.

[0031] Fig. 6A shows a case where A1 and B2 are large, and Fig. Figure 6B shows a case where A1 and B2 are small. The total area of ​​the fitting ellipses in Fig. 6A is in relation to the total sum of the areas of the fitting ellipses in Fig. 6B about 1 / 10.

[0032] Furthermore, an arbitrary threshold value S is set with respect to the total area, and if the total area is greater than S, the correction stops. For example, if a value between the total area in the lower diagram of Fig. 6A and the total area in the lower diagram of Fig. 6B, where the threshold S is set, is used in the case of the lower diagram. Fig. 6A found that the correction according to the invention is necessary because the area is smaller than the threshold S; however, in the case of the lower diagram, Fig. 6B determined that the correction is not necessary because the area is larger than the threshold S.

[0033] Accordingly, if the total area is less than S, the focus is changed by Δf until it reaches nΔf to obtain the Ronchigrams, local autocorrelation functions are calculated, and the focus is then returned to its initial state (F9 to F12). A focal line at which the fitting ellipse extends furthest is detected for each local region on the side below the focus and on the side above the focus using the local autocorrelation functions of the through-focused Ronchigrams obtained in F2 to F12 (F13).

[0034] For example, a method for detecting the focal line is considered that uses the circularity of a fitting ellipse. If the radius (long diameter) of the fitting ellipse in the direction of the long axis is denoted as 'A' and the radius (short diameter) of it in the direction of the short axis is denoted as 'B', the circularity R is given by formula (6). R=2(a×b) / (a2+b2)

[0035] R takes on a maximum value of 1 in the case of a perfect circle and becomes small as the ellipse deviates from a perfect circle. Accordingly, if a graph is drawn where the magnitude of a focus is plotted on a transverse axis and R is plotted on a longitudinal axis, minima in the graph considered on the side below the focus and the side above the focus form the focal line.

[0036] The Fig. 7A and Fig. Figure 7B shows examples of focal line detection. The in Fig. Seven sections surrounded by circles represent focal lines that, under the second and third conditions from the left, are located on the side below the focal point. Fig. 7A were detected and also under the second and third conditions from the left on the side above the focal point in Fig. 7B were detected. Because the detection of the focal line here is done by comparing the relative dimensions of the fitting ellipses with different focal points for each local area, it is not necessary for the condition that the detected focal line be a true value.

[0037] For example, if, as in Fig. As shown in Figure 8(A), if the focal point spacing width Δf is small enough when through-focusing is performed, the true focal lines f1 and f2, at which the fitting ellipse extends furthest, can be detected, as shown in Figure 8(A). Fig. As shown in Figure 8(B), the focal lines f1' and f2', which are detected when the focal point separation width Δf is large, deviate from the true values. Accordingly, the values ​​|A1(τ)| and C1(τ) calculated using formulas (1) and (3) become approximate values, their accuracies being determined by Δf.

[0038] The procedure for using the local autocorrelation function in F6 to F13 has been described above; however, an absolute value of a Fourier transform in the local domain can be used instead. Because the values ​​to be calculated are C1, A1, and B2, formulas (7) and (8), obtained by approximating formula (4) and formula (5), are used, assuming that τ is small. C1(τ)≈C1+2(B2τ*+B2*τ) A1(τ)≈A1+2B2τ+2A2τ*

[0039] Because three equations are formed with respect to the real and imaginary parts of formulas (7) and (8) for each local region, it is necessary to detect the focal line in at least three local regions with respect to seven unknown quantities: C1, the real part ReA1 and the imaginary part ImA1 of A1, the real part ReB2 and the imaginary part ImB2 of B2, and the real part ReA2 and the imaginary part ImA2 of A2 (F14). If no detection occurs, the focal line is highlighted by introducing A1, and the through-the-focal-point region T is enlarged (F15) to detect the focal line again.

[0040] If the focal line is detected in three or more local regions, C1(τ) and A1(τ) (|A1(τ)| and ∠A1(τ)) are calculated in each local region using formulas (1) to (3) (F16). If the central coordinates τ in the local regions are as in Fig. As represented in 9, the simultaneous equation to be solved becomes formula (9). (1004u14v1000102u1−2v12u1−2v10012v12u1−2v12u11004u24v2000102u2−2v22u2−2v20012v12 u2−2v22u2•••••••••••••••••••••1004u94v9000102u9−2v92u9−2v90012v92u9−2v92u9)×(C1Re A1Im A1Re B2Im B2Re A2Im A2)=(C1(τ1)Re A1(τ1)Im A1(τ1) C1(τ2)Re A1(τ2)Im A1(τ2)•••C1(τ9)Re A1(τ9)Im A1(τ9))

[0041] Here, ReA1(τ) and ImA1(τ) are calculated using formula (10) and formula (11) based on |A1(τ) l and ∠A1(τ) = θ. ReA1(τ)=|A1(τ)|cos[2θ] ImA1(τ)=|A1(τ)|sin[2θ]

[0042] Formula (9) is solved using the method of least squares (F17), and the values ​​of A1 and B2 are compared with target values ​​(F18). If the values ​​are lower than the target values, the correction is stopped; however, if the values ​​are higher than the target values, C1, A1, and B2 are corrected (F19).

[0043] In this way, F2 to F19 are repeated until A1 and B2 become smaller than the target values. However, by decreasing the through-the-focal-point region T and the focal-point separation width Δf(F20) in each repetition, the detection accuracy of the focal line and the measurement accuracy of the aberration coefficient increase. For example, as a method for determining the through-the-focal-point region T, a procedure for calculating A1(τ) with respect to each local region using formula (8) with respect to A1, B2, and A2 calculated by the least-squares method, and setting T as twice the maximum of the calculated values ​​of A1(τ), may be considered.

[0044] With this method, the astigmatism difference can be expected to be maximal before A1 and B2 are corrected in (F19), and smaller than the maximum after correction. It is necessary to determine the focal distance Δf, taking measurement time and accuracy into account; however, a method for setting Δf to 1 / 10 T, for example, could be considered.

[0045] Because, as described above, three equations are formulated relating to the calculated unknown quantities, the focal line must also be detected in three or more local regions. Since the number of equations required depends on the type of aberration to be calculated, the detection of the focal line in three or more local regions is not essential.

[0046] Here, with reference to the correspondences of Fig. 1 and the Fig. 5A and Fig. 5B notes that the through-focal-point mapping unit 18 corresponds to F2, F3, F9, F11, and F12; the input image validity determination unit 19 corresponds to F4 and F5; the correction necessity determination unit 20 corresponds to F6 to F8; the off-axis aberration calculation unit 21 corresponds to F10 and F13 to F16; the on-axis aberration calculation unit 22 corresponds to F17 and F18; the correction control amount calculation unit 23 corresponds to F19, the conversion coefficient table 24 to F19, and the control unit 26 to F19; and the optimal focal area calculation unit 25 corresponds to F20. F1 is performed by a user through the input / output unit 15, but it can be omitted using predefined values.

[0047] Fig. Figure 10 shows an example of a GUI for automatic aberration correction. The following describes a procedure for performing automatic correction using the GUI. First, the through-focal area T is set using a through-focal area setting text box 28, and the focal distance width Δf is set using a focal distance width text box 29.

[0048] Then, by pressing a correction start key, 31 F2 to F20 will be entered into Fig. 5B is executed automatically. Captured Ronchigrams are displayed in a Ronchigram display 27 in Fig. 10 is displayed, and a focal point value at that time is shown in a focal point indicator marker 30. A status indicating whether the detection of the focal line was successful is shown in a status bar 32.

[0049] In the flowchart from the Fig. 5A and Fig.5B Formulas (7) and (8) are used with respect to off-axis defocus and off-axis astigmatism using C1, A1 and B2 as correction targets, by further calculating the off-axis defocus and off-axis astigmatism with respect to a local area where τ is large, using formula (4) and formula (5), considering aberrations up to the fifth order, or an equation, considering aberrations up to an even higher order, however, the invention can also be applied to the measurement of higher order aberrations.Furthermore, the invention is used as an effective means for correcting large values ​​of A1 and B2 that are not easily measured using known techniques. However, even when A1 and B2 are very small, requiring final fine-tuning, the invention can solve the problem by reducing the focal distance Δf. This means that the invention can first be used to correct large aberration values ​​(A1 and B2), and then the final fine-tuning can be performed using a different aberration correction device or method after the aberrations have become smaller than a certain threshold. In this way, several devices or methods can be combined as needed.Furthermore, as described above, the processes for final fine-tuning can be performed by a single aberration correction device or a single aberration correction procedure.

[0050] Above, an embodiment of the invention has been described as an example of the scanning electron microscope; however, the invention can also be applied to another device operating with a beam of charged particles and equipped with Ronchigram observation means, for example, a transmission electron microscope.

[0051] According to the invention, by extracting characteristic aberration values ​​from several Ronchigrams obtained by through-focusing, an automatic aberration correction can be performed even in a state where A1 and B2 remain with large values ​​that are not easily measured by known techniques. Reference symbol list 1 ELECTRON BEAM SOURCE 2 ELECTRON BEAM 3, 4 RADIATION LENS 5 ADJUSTING LENS 6 Aberration Corrector 7 TRANSFER LENS 8 SAMPLING COIL 9 LENS 10 SAMPLE 11 Projection Lens 12 Darkfield Image Detector 13 CAMERA 14 Darkfield Image Observation Unit 15 INPUT AND OUTPUT UNIT 16 RONCHIGRAM OBSERVATION UNIT 17 AUTOMATIC ABERRATION CORRECTION DEVICE 18 THROUGH-FOCUS IMAGE UNIT 19 Entrance Image Validity Unit 20 Correction Necessity Determination Unit 21 UNIT FOR CALCULATING OFF-AXIS ABERRATION 22 UNIT FOR CALCULATING ABERRATION ON THE AXIS 23 Correction Tax Amount Calculation Unit 24 Conversion Coefficiency Table 25 OPTIMAL FOCAL AREA CALCULATION UNIT 26 CONTROL UNIT 27 RONCHIGRAMM DISPLAY 28 THROUGH-THE-FOCUS-AREA DEFINITION TEXT BOXES 29 focal point distance text boxes 30 FOCAL POINT INDICATOR MARKER 31 CORRECTION START BUTTON 32 STATUS BAR

Claims

[1] A device operating with a beam of charged particles, comprising the following: a source (1) for a beam (2) of charged particles, an optical system operating with charged particles, designed to irradiate a sample (10) with charged particles emitted from the source (1) for a beam (2) of charged particles, an aberration corrector (6) designed to correct the aberration of the charged particle optical system, a control unit (26) designed to control the charged particle optical system and the aberration corrector (6), a through-the-focal-point imaging unit (18) designed to obtain, in a first step, several Ronchigrams in which the focal point position of the charged particle optical system is changed, an aberration calculation unit designed to subdivide the obtained Ronchigram into several local areas in a second step and to calculate the aberration magnitude based on focal lines detected in the local areas in a third step, and an input unit (15) designed for inputting a setting value, characterized by , that the through-focal-point imaging unit (18) is designed to change the focal point position within a range based on the entered setting value, The aberration calculation unit is designed to detect the focal lines in the local areas of each obtained Ronchigram on the side below the focal point and on the side above the focal point, and the through-focal-point imaging unit (18) is designed to obtain multiple Ronchigrams when the focal point position is changed in an area larger than the area based on the input setting value, when the focal lines detected in the local areas are smaller than a defined value. [2] Device operating with a beam of charged particles according to claim 1, wherein the aberration calculation unit is designed to fit isointensity lines in the local areas using ellipses and to detect the focal lines based on the ellipses for fitting. [3] Device operating with a beam of charged particles according to claim 2, wherein the aberration calculation unit is designed to calculate the isointensity lines in the local regions by means of an autocorrelation function or a Fourier transform. [4] Device operating with a beam of charged particles according to claim 1, wherein the optical system operating with a beam (2) of charged particles has an objective lens (9) which is designed to focus the charged particles onto the sample (10), and the several local regions include a first local region corresponding to a first electron beam flux passing through the vicinity of the axis of the optical system operating with a beam of charged particles, and a second local region corresponding to a second electron beam flux passing through the objective lens (9) further outwards with respect to the first electron beam flux. [5] A device operating with a beam of charged particles according to claim 1, comprising the following: a correction start input unit (31) designed to receive a start input of a process for obtaining the multiple Ronchigrams through the through-the-focal-point imaging unit (18), and a status indicator designed to show, based on the Ronchigrams received, whether the detection of focal lines is successful, wherein, if the detection of the focal lines is not performed, the aberration corrector (6) is designed to send a command to request an input of a setting value back to the input unit (15) or to send a command to obtain the multiple Ronchigrams in which the focal position has changed in an area larger than the area in which the focal position has changed to the through-the-focal-point imaging unit (18). [6] Method for correcting an aberration in a device operating with a beam of charged particles, comprising a source (1) for a beam (2) of charged particles, a charged particle optical system irradiating a sample (10) with charged particles emitted by the source (1) for a beam (2) of charged particles, an aberration corrector (6) correcting the aberration of the charged particle optical system, and a control unit (26) controlling the charged particle optical system and the aberration corrector (6), wherein the method comprises: a first step in obtaining several Ronchigrams in which the focal point position of the charged particle optical system is changed, a second step of dividing the obtained Ronchigrams into several local areas, a third step of calculating the aberration magnitude based on the focal lines detected in the local areas, and a step to enter a setting value to be used in the first step, characterized by , that The first step changes the focal point position within a range based on the entered setting value. The third step is a step of detecting the focal lines in the local areas of each obtained Ronchigram on the side below the focal point and the side above the focal point, and Furthermore, in the procedure, if the focal lines detected in the local areas are smaller than a defined value, the multiple Ronchigrams in which the focal point position is changed are obtained in an area that is larger than the area based on the input setting value. [7] Method for correcting the aberration according to claim 6, wherein the third step comprises a fourth step of fitting isointensity lines in the local areas using ellipses and a fifth step of detecting the focal lines based on the ellipses for fitting. [8] Method for correcting the aberration according to claim 7, wherein the isointensity lines in the local regions are calculated in the fourth step by an autocorrelation function or a Fourier transform.

Images