Method of determining aberrations of a charged particle beam and charged particle beam system

Abstract

A method is described for determining the aberrations of a charged particle beam (11) focused onto a sample (10) by a focusing lens (120) having a given numerical aperture (NA) in a charged particle beam system. The method includes: (a.) determining the aberrations of a set of beam aberration coefficients (C) based at least on the given numerical aperture (NA). 1...n (a) Simulate one or more beam cross sections under one or more first defocus settings for each of two or more distinct values ​​of the first beam aberration coefficient (C1) in the sample to provide a plurality of first simulated beam cross sections; (b) Extract two or more values ​​of the first aberration characteristic (~C1) associated with the first beam aberration coefficient (C1) from the plurality of first simulated beam cross sections; (c) Determine a first correlation between the first beam aberration coefficient (C1) and the first aberration characteristic (~C1); (d) Capture one or more images of the sample under one or more first defocus settings or under one or more second defocus settings to provide one or more captured images, and retrieve one or more retrieved beam cross sections from the one or more captured images; (e) Extract the retrieved value of the first aberration characteristic from the one or more retrieved beam cross sections; and (f) Determine the actual value of the first beam aberration coefficient based on the first correlation and based on the retrieved value of the first aberration characteristic.

Description

Technical Field

[0001] The embodiments described herein relate to methods for determining aberrations of a charged particle beam in a charged particle beam system, such as in electron microscopy, particularly in scanning electron microscopy (SEM). Specifically, actual values ​​of one or more beam aberration coefficients can be determined to facilitate beam aberration correction and improve resolution. More specifically, the embodiments described herein relate to methods for determining the beam aberration coefficients of a charged particle beam focused onto a sample by a focusing lens having a given numerical aperture in a charged particle beam system, which helps to provide an aberration-corrected charged particle beam. The embodiments also relate to charged particle beam systems for examining and / or imaging samples, said charged particle beam systems configured for any of the methods described herein. Background Technology

[0002] Modern semiconductor technology places high demands on the structuring and detection of samples at the nanometer and even sub-nanometer scale. Micrometer- and nanometer-scale process control, inspection, or structuring is typically performed using charged particle beams (e.g., electron beams), which are generated, shaped, deflected, and focused in charged particle beam systems such as electron microscopes or electron beam pattern generators. For inspection purposes, charged particle beams offer significantly higher spatial resolution compared to, for example, photon beams.

[0003] Inspection equipment using charged particle beams (such as scanning electron microscopes (SEMs)) has numerous functions across various industrial sectors, including but not limited to electronic circuit inspection, exposure systems for photolithography, inspection systems, defect inspection tools, and testing systems for integrated circuits. In such particle beam systems, fine-beam probes with high current densities can be used. For example, in the case of SEMs, the primary electron beam generates signal particles, such as secondary electrons (SE) and / or backscattered electrons (BSE), that can be used to image and / or inspect samples.

[0004] However, reliably examining and / or imaging samples with good resolution using charged particle beam systems is challenging because charged particle beams are typically affected by beam aberrations that limit the achievable resolution. In typical charged particle beam systems, aberration correctors are provided to at least partially compensate for aberrations in the charged particle beam, such as spherical aberration, astigmatism, and / or chromatic aberration. Compared to an uncorrected beam, an aberration-corrected charged particle beam can provide a smaller detection focal length, thus offering better resolution. However, properly tuning the settings of aberration correctors is challenging, as these correctors may have numerous controls to satisfactorily correct beam aberrations because the beam aberrations present in the system are often unknown.

[0005] Several methods for determining or correcting aberrations of charged particle beams in electron microscopy have been described. For example, Uno et al., in their paper "Aberration correction and its automatic control in scanning electron microscopes" (hereinafter referred to as "UNO") published in the *Optik-International Journal for Light and Electron Optics*, 2005, 116(9), pp. 438-448, describe a method for determining the correlation between so-called "digital aberrations" and the field strength of a specific multilevel aberration corrector adapted to compensate for such "digital aberrations." However, UNO's aberration correction method is time-consuming because multiple measurements must be performed to determine the aforementioned correlation, and the use of the determined correlation is limited to a specific multilevel aberration corrector adapted to generate the corresponding field strength.

[0006] In light of the foregoing, it would be beneficial to provide a rapid and reliable method for determining the aberrations of a charged particle beam focused by a focusing lens in a charged particle beam system, particularly for determining the actual values ​​of the beam aberration coefficients, i.e., the absolute values ​​of the beam aberration coefficients. Furthermore, it would be beneficial to provide a charged particle beam system for examining and / or imaging samples, configured to operate according to any of the methods described herein. The determined actual aberration values ​​can be used, for example, to correct the charged particle beam and / or to examine the beam spot shape of the actual charged particle beam, for example, to compare the performance of different charged particle beam tools. Summary of the Invention

[0007] In view of the foregoing, a method for determining the beam aberration of a charged particle beam and a charged particle beam system configured for determining the beam aberration of a charged particle beam are provided according to the independent claims.

[0008] According to one aspect, a method is proposed for determining the aberrations of a charged particle beam focused onto a sample by a focusing lens having a given numerical aperture in a charged particle beam system. The method includes: (a.) simulating one or more beam cross sections in one or more first defocus settings, at least based on a given numerical aperture, for each of two or more distinct values ​​of a first beam aberration coefficient in a set of beam aberration coefficients, to provide a plurality of first simulated beam cross sections; (b.) extracting two or more values ​​of a first aberration characteristic associated with the first beam aberration coefficient from the plurality of first simulated beam cross sections; (c.) determining a first correlation between the first beam aberration coefficient and the first aberration characteristic; (d.) capturing one or more images of the sample in one or more first defocus settings or in one or more second defocus settings to provide one or more captured images, and retrieving one or more retrieved beam cross sections from the one or more captured images; (e.) extracting retrieved values ​​of the first aberration characteristic from the one or more retrieved beam cross sections; and (f.) determining the actual value of the first beam aberration coefficient based on the first correlation and based on the retrieved values ​​of the first aberration characteristic.

[0009] According to one aspect, a method is proposed for determining the aberrations of a charged particle beam focused onto a sample by a focusing lens having a given numerical aperture in a charged particle beam system. The method includes: (a.) simulating one or more beam cross sections in one or more first defocus settings, at least based on the given numerical aperture, for each of two or more distinct values ​​of a first beam aberration coefficient in a set of beam aberration coefficients, to provide a plurality of first simulated beam cross sections; (b.) extracting two or more values ​​of a first aberration characteristic associated with the first beam aberration coefficient from the plurality of first simulated beam cross sections; and (c.) determining a first correlation between the first beam aberration coefficient and the first aberration characteristic for determining the actual value of the first beam aberration coefficient in the charged particle beam focused onto the sample using the given numerical aperture. Specifically, the first correlation can be determined and subsequently stored in a memory associated with a processing unit of the charged particle beam system for subsequent determination of the actual value of the first beam aberration coefficient.

[0010] In some embodiments, one or more first defocusing settings include one or more defocusing distances from a corresponding focal point of the charged particle beam, particularly including a pair of defocusing distances comprising an underfocusing distance and a corresponding overfocusing distance, at which one or more beam cross-sections are simulated in (a.) and / or one or more images are captured in (d.). Alternatively or additionally, one or more first defocusing settings include one or more beam landing energies of the charged particle beam varying from the focused beam landing energy, at which one or more beam landing energies are simulated in (a.) and / or one or more images are captured in (d.).

[0011] According to one aspect, a charged particle beam system is provided for imaging and / or inspecting a sample using a charged particle beam, particularly an electron beam. The charged particle beam system includes: a charged particle source for emitting a charged particle beam propagating along an optical axis; a sample platform; a focusing lens for focusing the charged particle beam onto a sample placed on the sample platform using a given numerical aperture; a charged particle detector for detecting signal particles emitted from the sample; and a processing unit and a memory storing a first correlation between first beam aberration coefficients and a first aberration characteristic, the first aberration characteristic being a parameter extractable from one or more defocused beam cross-sections and having a predetermined relationship with the first beam aberration coefficients.

[0012] According to one aspect, a charged particle beam system is described, the charged particle beam system being configured to focus a charged particle beam having a given numerical aperture onto a sample, particularly according to the foregoing aspect. The charged particle beam system includes a simulation unit configured to: (i) simulate one or more beam cross sections in one or more defocusing settings for each of two or more distinct values ​​of a first beam aberration coefficient to provide a plurality of first simulated beam cross sections; (ii) extract two or more values ​​of a first aberration characteristic from the plurality of first simulated beam cross sections; and (iii) determine a first correlation between the first beam aberration coefficient and the first aberration characteristic to determine the actual value of the first beam aberration coefficient in the charged particle beam focused onto the sample using the given numerical aperture. The determined first correlation may optionally be stored in memory. The charged particle beam system may also include any of the features of the charged particle beam system described herein and / or may be configured to perform any of the methods described herein.

[0013] The embodiments also relate to apparatus for performing the disclosed methods and include apparatus components for performing various method actions. The methods can be performed by hardware components, a computer programmed with suitable software, any combination of both, or in any other manner. Furthermore, the embodiments also relate to methods of operating the described apparatus.

[0014] Further advantages, features, aspects and details that can be combined with the embodiments described herein will become apparent from the dependent claims, the specification and the drawings. Attached Figure Description

[0015] To gain a more detailed understanding of the features described above, a more specific description summarized above can be obtained by referring to the embodiments. The accompanying drawings relate to one or more embodiments and are described below.

[0016] Figure 1A schematic diagram of a charged particle beam system according to an embodiment described herein is shown, the charged particle beam system being adapted to operate according to any of the methods described herein;

[0017] Figure 2 A diagram illustrating a method for determining a first correlation (f1) between aberration coefficients and corresponding aberration characteristics according to embodiments described herein, the first correlation (f1) being used to determine the aberrations of a charged particle beam; and

[0018] Figure 3 A diagram illustrates a method for determining the aberrations of a charged particle beam in a charged particle beam system according to embodiments described herein. Specific Implementation

[0019] Reference will now be made in detail to various embodiments, one or more examples of which are illustrated in the accompanying drawings. In the following description of the drawings, the same reference numerals refer to the same parts. Generally, only differences with respect to the various embodiments are described. Each example is provided by way of explanation and is not intended to be limiting. Furthermore, features shown or described as part of one embodiment may be used in other embodiments or in combination with other embodiments to produce further embodiments. This specification is intended to include such modifications and variations.

[0020] Figure 1 This is a schematic diagram of a charged particle beam system 100 for examining and / or imaging a sample 10 according to embodiments described herein. The charged particle beam system 100 includes a charged particle source 105, particularly an electron source, for emitting a charged particle beam 11, particularly an electron beam, propagating along optical axis A. The charged particle beam system 100 further includes a sample platform 108 and a focusing lens 120, particularly an objective lens, for focusing the charged particle beam 11 onto the sample 10 placed on the sample platform 108. The charged particle beam system 100 also includes a charged particle detector 118, particularly an electron detector, for detecting signal particles (particularly secondary electrons and / or backscattered electrons) emitted from the sample 10.

[0021] An image generation unit 160 may be provided, configured to generate one or more images of the sample 10 based on charged particle signals received from the charged particle detector 118. The image generation unit 160 may forward one or more images of the sample to a processing unit 170, which is configured to determine actual values ​​of the beam aberration coefficients therefrom according to the method described herein.

[0022] The sample platform 108 can be a movable platform. Specifically, the sample platform 108 can move in the Z direction, i.e., in the direction of the optical axis A, so that the distance between the focusing lens 120 and the sample platform 108 can be varied (see...). Figure 1 (Arrow 112 in the image). By moving the sample platform 108 in the Z direction, the sample 10 can be moved to the focusing plane p at a distance from the focusing lens 120. F Different "defocus distances" allow for the capture of defocused images of sample 10 by moving the platform accordingly, for example, in increments of 0.1 μm or greater and / or 2 μm or less. In some embodiments, sample platform 108 can also be moved in a plane perpendicular to the optical axis A (also referred to herein as the XY plane). By moving sample platform 108 within the XY plane, specific surface regions of sample 10 can be moved to a region below focusing lens 120, allowing the specific surface region to be imaged by focusing the charged particle beam 11 onto that specific surface region.

[0023] The beam optics of the charged particle beam system 100 are typically housed in a vacuum chamber 101 that can be evacuated, allowing the charged particle beam 11 to propagate along optical axis A from the charged particle source 105 toward the sample platform 108 and impact the sample 10 at sub-atmospheric pressure, for example, at pressures below 1000 kcal / kg. -3 millibars or pressure below 10 -5 In the case of millibars.

[0024] In some embodiments, the charged particle beam system 100 may be an electron microscope, particularly a scanning electron microscope. A scanning deflector 107 may be provided for scanning the charged particle beam 11 on the surface of the sample 10 along a predetermined scanning pattern, for example, along the X and / or Y directions.

[0025] In some embodiments, the condenser lens system 106 may be arranged downstream of the charged particle source 105, particularly for collimating the charged particle beam 11 propagating toward the focusing lens 120. In some embodiments, the focusing lens 120 is an objective lens configured to focus the charged particle beam 11 onto the sample 10, particularly a magnetic objective lens, an electrostatic lens, or a combination of magnetic and electrostatic lenses.

[0026] One or more surface areas of sample 10 may be inspected and / or imaged using charged particle beam system 100. As used herein, the term “sample” may refer to a substrate, such as a substrate on which one or more layers or features are formed, a semiconductor wafer, a glass substrate, a flexible substrate (such as a mesh substrate), or another sample to be inspected. The sample may be inspected for one or more of the following: (1) imaging the surface of the sample, (2) measuring the dimensions of one or more features of the sample, for example in the lateral direction, i.e., in the XY plane, (3) performing critical dimension measurements and / or metrology, (4) detecting defects, and / or (5) investigating the quality of the sample.

[0027] To examine sample 10 using charged particle beam 11, charged particle beam 11 is typically focused onto the sample surface using focusing lens 120. When charged particle beam 11 strikes the sample surface, secondary electrons and / or backscattered electrons (referred to as "signal electrons") are emitted from the sample. The signal electrons provide information about the spatial characteristics and dimensions of the sample features and can be detected using charged particle detector 118. By scanning the charged particle beam 11 across the sample surface, for example, using scanning deflector 107, and detecting the signal electrons as a function of the generation location of the signal electrons, the sample surface or a portion thereof can be imaged, for example, using image generation unit 160, which can be configured to provide an image of sample 10 based on the received signal electrons.

[0028] The small spot of the focused charged particle beam 11 on the sample surface increases the achievable image resolution. Therefore, during inspection, the sample surface should be positioned on the focal plane p of the focusing lens 120. F Within this scope, a clear, focused image of sample 10 is obtained. The clear image of sample 10 captured in this document is also referred to as the "focus image". F The subscript F indicates "focus," while images taken out of focus are referred to as defocused images in this article, such as overfocused images taken behind the focal point. o-1 Or an underfocused image taken in front of the beam focus h u-1 Similarly, the charged particle beam 11 is focused on the plane p. F The beam cross section in this paper is referred to as the "focused beam cross section g". F The defocused beam cross section, referred to in this paper as the defocused beam cross section, is such as the overfocused beam cross section g behind the beam focal point. o-1 Or the underfocused beam cross-section g in front of the beam focal point u-1 .

[0029] It is worth noting that images can be mathematically represented in real space (equivalent to in the image domain, i.e., as a function of spatial coordinates) or in Fourier space (equivalent to in the frequency domain, i.e., as a function of spatial frequency). An image in Fourier space can be computed from an image in real space using the Fourier transform (FT). Both representations contain the relevant information about the image. As used in this paper, an image in real space is represented by the lowercase letter "h". n The figure in Fourier space is represented by the capital letter "H". n " indicates. For example, "h F "H" represents the focused image of the sample in real space. F "H" represents the focused image of the sample in Fourier space. F It is h F The Fourier transform of . Similarly, the beam cross-section in real space is represented in this paper by the lowercase letter "g".n The term "G" indicates that the cross-section of the beam in Fourier space is represented by the capital letter "G" in this paper. n " indicates. For example, "g F "G" represents the focused cross-section of a beam of charged particles in real space. F "G" represents the focused cross-section of a beam of charged particles in Fourier space. F It is g F The Fourier transform of images and beam cross-sections in real space can be Fourier transformed to Fourier space and vice versa using the Fast Fourier Transform (FFT) algorithm in some embodiments described herein.

[0030] In charged particle beam systems, beam aberrations typically cause beam cross-section enlargement or distortion, which reduces the achievable resolution. For example, spherical aberration, often introduced by lenses in the system, can cause the focal plane p to... F The increased cross-section of the focusing beam in the image can cause astigmatism, which may result in light rays propagating in different planes having different focal points, thus blurring the image.

[0031] Different types of beam aberrations may exist in charged particle beam systems and may require correction, such as, for example, (1) spherical aberration (determined by the beam aberration coefficient C). 3，0 Or C s (Quantitative representation), (2) Defocus (by the beam aberration coefficient C) df (3) First-order astigmatism (quantitatively represented by the beam aberration coefficient A1), (4) Second-order astigmatism (quantitatively represented by the beam aberration coefficient A2), (5) Third-order astigmatism (quantitatively represented by the beam aberration coefficient A3), (6) Coma (quantitatively represented by the beam aberration coefficient B2), (7) Astral aberration (quantitatively represented by the beam aberration coefficient S3).

[0032] Furthermore, due to the energy diffusion of the charged particle beam and the dispersion of the system's beam optics, multiple chromatic aberrations may exist, which can be quantitatively represented by one or more chromatic aberration coefficients. A set of beam aberration coefficients C 1…n This can include two, three, or more of the aforementioned beam aberration coefficients, for example, C. 1…n =[C df C s [A1, A2]. According to the embodiments and methods described herein, a set of beam aberration coefficients C can be determined for an actual charged particle beam existing in a charged particle beam system. 1…n The absolute value of the beam aberration coefficients allows beam aberrations to be compensated using one or more beam aberration correctors adjusted according to the determined absolute value.

[0033] Beam aberration can be corrected using an aberration corrector, such as an electrostatic or magnetic multipole corrector. Aberration corrector 109 is... Figure 1The diagram is schematic, but it should be understood that a charged particle beam system may also include two or more aberration correctors, which do not necessarily have to be located at a single position along the optical axis A. For example, a quadrupole astigmatism correction device may be provided to correct first-order astigmatism A1, and / or higher-order multipole devices may be provided to correct higher-order astigmatism A2 and / or A3. Additional correctors may be provided to compensate for spherical aberration C. s Various types of aberration correctors are known for correcting various bundle aberration coefficients.

[0034] Adjusting aberration correctors to properly correct one or more types of beam aberrations is challenging because the amount of beam aberration present in a system is often unknown. One or more aberration correctors can be configured to compensate for pre-calculated beam aberrations theoretically introduced by the system's beam optics, but this approach is often not accurate enough. Specifically, not all sources of beam aberration are known, especially in terms of quantity. For example, beam aberrations can also be introduced by inaccuracies in the system, such as mechanical, magnetic, or electrostatic inaccuracies, charge contamination, material inhomogeneities, and manufacturing defects, which are initially unknown. Beam optics can include one or more of objectives, collimators, deflectors, scanning deflectors, beam splitters, charged particle detectors, and aberration correctors.

[0035] Various methods can be used to estimate beam aberrations in charged particle beam systems, allowing for at least partial correction of these aberrations. Some methods rely on visual inspection of the charged particle beam, such as in the far-field case, which can indicate the aberrations present in the system. Other methods rely on analysis of captured images. Defocused images produce information about the corresponding beam cross-section (i.e., probe shape) at the defocus point, and the defocused probe shape can provide qualitative information about specific types of beam aberrations. For example, astigmatic beams are often non-rotationally symmetric, especially far from the beam focus. Methods rely on extracting line profiles from the defocused beam cross-section, from which beam aberrations can be estimated.

[0036] UNO describes the extraction of beam cross sections (i.e., probe shapes) from defocused images in Sections 6 and 6.1, and the determination of “digital aberrations” from underfocused and overfocused probe shapes in Section 6.2. Digital aberrations can be retrieved from overfocused and underfocused beam cross sections and can be defined for different beam aberrations, see UNO’s formulas (38)-(47). UNO incorporates by reference hereof its determination of “digital aberrations” (referred to herein as “aberration characteristics”) from defocused beam cross sections and its determination of defocused beam cross sections (particularly underfocused and overfocused beam cross sections) from defocused images (UNO, Chapter 6).

[0037] According to UNO, multiple measurements are performed to determine the correlation between digitized aberrations and the correction field applied by a specific bundle aberration corrector. The determined correlation is then correlated with a specific multi-level aberration corrector suitable for generating the corresponding field strength, which makes the UNO method time-consuming and limits its application areas.

[0038] The above concepts only allow for relative estimations of aberrations, not for retrieving the absolute values ​​of aberration coefficients. The "absolute value" of a bundle aberration coefficient can be understood as the actual quantitative value of the bundle aberration coefficient, such as C expressed in [mm]. s These values ​​allow for the appropriate setting of the aberration corrector directly based on the absolute values ​​of the determined bundle aberration coefficients. It is worth noting that previously known methods typically only achieve relative estimates of bundle aberrations, which may vary from measurement and / or be related to the specific correction field generated by a particular aberration corrector.

[0039] The method described herein allows for the accurate and reliable determination of actual values ​​of beam aberration coefficients that quantitatively describe the beam aberrations of a charged particle beam. In some embodiments, the actual values ​​of the beam aberration coefficients can be determined by processing unit 170 and then directly forwarded to aberration corrector 109, such that aberration corrector 109 can compensate for one or more beam aberrations and provide an aberration-compensated charged particle beam.

[0040] The method described herein uses a calibration based on a simulation of the beam profile defocused at a predetermined defocus setting to determine the correlation between beam aberration coefficients and corresponding aberration characteristics. For example, a first correlation f1 between a first beam aberration coefficient C1 (absolute value) and its corresponding first aberration characteristic ~C1 is determined based on the simulated first calibration. The first correlation f1 can optionally be stored in the memory 171 of the charged particle beam system, such that the calibration is readily available and can be used to rapidly convert the values ​​of the first aberration characteristics retrieved from one or more defocused images into actual values ​​of the first aberration coefficients (expressed in absolute value, such as those used for aberration correction in an aberration corrector). The following explains how the correlation f1 between the beam aberration coefficients and corresponding aberration characteristics is determined according to the method described herein. 1...n .

[0041] Figure 2 The first aberration coefficient C1 is determined in an exemplary manner (here: C1 = C s A diagram illustrating the method for determining the first correlation f1 between the spherical aberration coefficient (i.e., the spherical aberration coefficient) and the corresponding first aberration characteristic ~ C1, whereby the first correlation f1 is used to determine the aberration of a charged particle beam (here: the absolute value of the spherical aberration is determined by the spherical beam aberration coefficient C1). s (The actual value is represented). It is worth noting that, for illustrative purposes, Figure 2The beam cross section shown is simulated using other non-zero beam aberration coefficients (such as non-zero astigmatism), such that the beam cross section is not rotationally symmetric, and the radial line profile p o-1 and p u-1 These depend on the azimuth angle, respectively. However, those skilled in the art will understand that the embodiments described herein include simulating, in step (a.), only one non-zero aberration coefficient (e.g., C). s ≠0) beam cross sections, or alternatively simulate beams with several non-zero aberration coefficients (e.g., C s The cross-section of the bundle (≠0 and A1≠0).

[0042] In box 210, for two or more different values ​​of the first aberration coefficient C1 ( x1 C1, x2 Each of C1) simulates one or more beam cross sections (g) under one or more first defocus settings. o-1 g u-1 (Step (a.)) to provide multiple simulated beam cross sections (g o-1 g u-1 The two or more distinct values ​​of the aberration coefficients of the first bundle can be arbitrarily chosen (e.g., Figure 2 In the example x1 C1 = -0.8 mm and x2 C1 = +1.2 mm), for example, the negative and positive values ​​corresponding to charged particle beams with negative and positive spherical aberrations, especially the value range from a typical uncorrected system.

[0043] Simulations can be performed for a given value of the numerical aperture (NA) of the charged particle beam, where the simulated NA essentially corresponds to the NA of the actual charged particle beam system whose aberrations are to be determined. The numerical aperture (NA) is a dimension known to those skilled in the art, characterizing the divergence of the charged particle beam toward the beam focus provided by the focusing lens 120. The values ​​of the aberration characteristics determined under a specific defocus setting can depend on the numerical aperture, i.e., on the beam divergence toward the beam focus, making it advantageous to use the NA of the actual charged particle beam system for simulations.

[0044] In some embodiments, the beam landing energy E of the charged particles in the charged particle beam on the sample can be measured. L The simulation is performed using a given value, and the landing energy E used in the simulation is... L The beam landing energy E of the charged particle beam corresponding to the actual charged particle beam system with the aberration to be determined LThe beam landing energy can correspond to the (average) energy of the electrons impacting the sample and can be expressed in eV. The aberration characteristic values ​​determined under a specific defocus setting may depend on the charged particle energy, making the beam landing energy Ea of a practical charged particle beam system... L Conducting simulations may be beneficial.

[0045] In some embodiments, one or more first defocusing settings may include one or more defocusing distances (z) from the focal point of the charged particle beam. o-1 , z u-1 ), such that one or more beam cross sections (g o-1 ,g u-1 At one or more defocus distances (z o-1 , z u-1 Simulations are performed on the focal plane. Specifically, beam cross-sections are simulated at positions at certain positive and / or negative distances (z≠0) from their respective focal points, such that multiple first simulated beam cross-sections are defocused beam cross-sections. If beam aberrations exist in the charged particle beam, these aberrations are more pronounced at the defocus distance than at the focal plane, for example, because the focal spot is very small. Therefore, aberration characteristics can be retrieved more reliably and accurately from the defocused beam cross-section compared to the focused beam cross-section. One or more defocus distances may include a pair of defocus distances, namely, the overfocus distance (z≠0). o-1 ) and underfocus distance (z u-1 In other words, the beam cross-section can be simulated on both sides of the beam focal point, especially at the corresponding positive and negative distances from the beam focal point, referred to in this paper as the overfocus distance and underfocus distance.

[0046] Specifically, in (a.), two or more distinct values ​​of the aberration coefficients of the first beam can be considered. xl C1, x2 Each of C1) is used to simulate the overfocused beam cross section and the underfocused beam cross section to provide multiple first simulated beam cross sections. Figure 1 An example is shown of the overfocus distance (z) at the beam focal point. o-1 Simulated cross-section of the focused beam at (g) o-1 ) and the underfocusing distance (z) at the beam focal point u-1 Simulated underfocused beam cross section (g) at location u-1 (In particular, using the given NA and the given E respectively) L The beam cross-section shown in the diagram is the first value of the first beam aberration coefficient. X1 C1) simulation, i.e., having strong spherical aberration (C s>0), which can be clearly identified by the difference in beam profile between the overfocused and underfocused beam cross-sections. As mentioned above, for illustrative purposes, other non-zero beam aberrations (e.g., A1 ≠ 0) were also used to simulate the illustrated beam cross-section, presenting... Figure 2 The non-rotationally symmetric beam cross section. Similarly, using the second value ( x2 C1) and optionally with a value different from the first value ( x1 Other values ​​of the first beam aberration coefficients of C1) are used to simulate the overfocused beam cross section and the underfocused beam cross section to provide multiple first simulated beam cross sections.

[0047] Alternatively or additionally, one or more first defocusing settings may include landing energy E from the focused beam. L Landing energies of one or more beams of varying charged particles. Landing energy E of a focused beam. L This can be understood as the beam landing energy of a real charged particle beam system that provides beam focusing on a sample, such that the sample is positioned at the focal point, and variations in the beam landing energy away from the focal point result in a defocusing setting (“energy defocusing”), i.e., the beam focal point moves to another plane. More specifically, the chromatic aberration of the charged particle beam system transforms energy defocusing into spatial defocusing. Therefore, varying beam landing energy results in a simulated beam cross-section in a plane away from the corresponding focal plane of the charged particle beam, i.e., a simulated defocused beam cross-section. The defocusing effect is related to the system’s (multiple) chromatic aberrations. Therefore, if the first beam aberration coefficient is a chromatic beam aberration coefficient, it is particularly advantageous to simulate the beam cross-section at one or more different beam landing energies; and if the first beam aberration coefficient is a beam geometric aberration coefficient, it is particularly advantageous to simulate the beam cross-section at one or more defocusing distances.

[0048] In boxes 220 and 230, (step (b.)) two or more values ​​of the first aberration characteristic ~C1 associated with the first aberration coefficient C1 are extracted from multiple first analog beam cross sections. x1 C1, ~ x2 C1).

[0049] As used herein, "aberration characteristics" are parameters that are individually related to one of the beam aberration coefficients and can be extracted from the defocused beam cross-section of a charged particle beam, particularly from the overfocused and underfocused beam cross-sections. Under specific defocusing settings (for a given NA and E at a specific distance from the beam focus...), LThere is usually a defined one-to-one correspondence between the beam aberration coefficients and the corresponding aberration characteristics. For example, aberration characteristics can be defined according to Section 6.2 of the UNO, “Digital Aberrations” (see Equations (38)-(47)), but this disclosure is not limited thereto. For example, aberration characteristics may have additional factors, or may be calculated separately from interpolation or averaging of several pairs of underfocused beam cross sections and overfocused beam cross sections. At least some of the aberration characteristics may also be defined based on other defocus settings, i.e., it is not necessary to use an overfocused beam cross section and an underfocused beam cross section for retrieval. Specifically, aberration characteristics are defined such that the values ​​of the aberration characteristics can be retrieved from one or more defocused beam cross sections of a charged particle beam, and such that there is a relationship (in particular a one-to-one correspondence within the range of interest) between the beam aberration coefficients and the corresponding aberration characteristics, provided that at least some parameters (such as NA, E) are preserved. L and / or acquire or simulate the defocusing settings of the beam cross-section).

[0050] In some embodiments, two or more values ​​of the first aberration characteristic ~C1 (~ xl C1, ~ x2 Each of C1) is extracted from the overfocused beam cross section and the corresponding underfocused beam cross section. For example, in Figure 2 In the depicted embodiment, respectively from the first value of the first beam aberration coefficient ( x1 C1) Simulated overfocused beam cross section g o-1 and the corresponding underfocused beam cross section g u-1 Extract the first value (~ x1 C1). Therefore, the first correlation f1 will determine the first value of the first bundle aberration coefficient ( x1 C1) is mapped to the first value of the first aberration characteristic (~ x1 C1). The second value of the first aberration characteristic (~ x2 C1) can be similarly determined such that at least two pairs of values ​​for the first correlation f1:C1-->~C1 are determined in step (b.).

[0051] In some embodiments that can be combined with other embodiments described herein, one, two or more radial line profiles are determined based on each of a plurality of first simulated beam cross sections, and two or more values ​​of a first aberration characteristic are calculated based on the radial line profiles. As used herein, "radial line profile" refers to a radial intensity profile that can be extracted from a beam cross section, particularly the beam spot intensity along a particular azimuth angle, which can be obtained, for example, by two-dimensional interpolation or by averaging the azimuth angle.

[0052] Figure 2 Box 220 in the diagram illustrates the determination of radial line profiles based on each of a plurality of first simulated beam cross-sections. For example, one or more radial line profiles po-1 (Here: four radial line profiles at four different angles from the center of the corresponding beam profile) can be obtained from the cross-section g of the focused beam. o-1 The middle is determined, and one or more radial line profiles p o-1 (Here: four radial line profiles at four different angles from the corresponding beam profile center) can be obtained from the underfocused beam cross-section g u-1 Determined. The first aberration characteristic can be calculated from one or more radial line profiles defined by a pair of overfocused and underfocused beam profiles.

[0053] Specifically, N radial line profiles can be derived from each of multiple simulated beam cross-sections surrounding the center of the corresponding simulated beam cross-section at a constant angular step θ. k =k is determined in the range k = 0...(N-1), where N can be a positive integer arbitrarily chosen based on the symmetry of the aberration coefficients to be analyzed. Figure 2 In an exemplary embodiment, N = 4, such that θ1 = 0°, θ2 = 45°, θ3 = 90°, and θ4 = 135°. In other embodiments, N can be less than or greater than 4. For example, if C1 = C df If N can be set to 1, then the bundle profile is rotationally symmetric in the case of defocus aberration.

[0054] In some embodiments that can be combined with other embodiments described herein, any one or more of the bundle width value (σ), asymmetry value (μ), and curvature value (ρ) can be retrieved from each of one, two, or more radial line profiles, and two or more values ​​of the first aberration characteristic can be calculated based on any one or more of the bundle width value (σ), asymmetry value (μ), and / or curvature value (ρ). Specifically, the bundle width value (σ), asymmetry value (μ), and / or curvature value (ρ) can be calculated according to the following formula:

[0055]

[0056]

[0057]

[0058] Where j represents the radial position in the corresponding line profile (j = 0 corresponds to the center of the bundle profile), pj represents the intensity at radial position j, and and

[0059] exist Figure 2 The described embodiment (where C1 = C sIn this context, the curvature value can be calculated for each of the radial line profiles of each previously determined underfocused beam cross section and overfocused beam cross section, and two or more values ​​of the first aberration characteristic can be calculated based on the curvature value according to the following formula (~ x1 C1, ~ x2 C1):

[0060]

[0061] The subscripts “u” and “o” represent the underfocused beam cross section and the corresponding overfocused beam cross section, respectively, and the subscript “k” represents the k-th line profile among the total N line profiles of each beam cross section.

[0062] In some embodiments, two or more values ​​of the first aberration characteristic (~ x1 C1, ~ x2 C1) can be obtained through the following formula (see...) Figure 2 One of the boxes in frame 230 is determined:

[0063] If C1 = C df but

[0064] If C1 = A1, then

[0065] If C1 = B2, then

[0066] If C1 = A2, then

[0067] If C1 = C s but

[0068] If C1 = S3, then

[0069] If C1 = A3, then The subscripts "u" and "o" represent the underfocused beam cross section and the corresponding overfocused beam cross section, respectively, and the subscript "k" represents the k-th line profile. However, as mentioned above, the definition of aberration characteristics is not limited to this, and other equivalent definitions can be used.

[0070] In determining two or more different values ​​of the aberration coefficient of the first beam ( xl C1, x2 C1) Two or more values ​​of the first aberration characteristic associated with it (~ xl C1, ~ x2 After C1), the first correlation f1 between the first aberration coefficient C1 and the first aberration characteristic ~C1 is determined (step (c.)). Figure 2 As shown in box 240.

[0071] In some embodiments that can be combined with other embodiments described herein, the first correlation f1 between the first aberration coefficient C1 and the first aberration characteristic ~C1 is a substantially linear correlation, characterized by a scaling factor (corresponding to Figure 2 The first correlation f1 is determined by the slope (or inverse slope of the corresponding inverse function) in the linear graph, and step (c.) may include determining the scaling factor. Furthermore, the y-intercept of the graph can be determined, thereby fully defining the first correlation f1. Determining the first correlation may specifically include determining (and optionally storing in memory) the scaling factor corresponding to the calibration factor between the first beam aberration coefficient and the first aberration characteristic. Since the linear correlation corresponds to a one-to-one function, if the retrieved value of the first aberration characteristic ~C1 retrieved from the actual beam profile is available, knowing the first correlation f1, and in particular knowing the slope (or inverse slope) or scaling factor, allows the determination of the actual value of the first aberration coefficient C1.

[0072] If the first correlation f1 is not perfectly linear or substantially nonlinear, more than two value pairs, each including the first aberration coefficient and the corresponding first aberration characteristic, can be retrieved by simulations similar to the two value pairs described above. This allows nonlinear correlations to be determined with a predetermined precision, which may depend on the number of value pairs retrieved by the corresponding simulations.

[0073] In some embodiments that can be combined with other embodiments described herein, the first correlation f1 may be stored in the memory of the charged particle beam system. Specifically, the first correlation f1 may be a substantially linear correlation characterized by a first scaling factor, and at least the first scaling factor may be stored in the memory of the charged particle beam system. The linear correlation may (later) be used to determine the actual value of the first beam aberration coefficient based on a retrieved value of a first aberration characteristic.

[0074] Back Figure 1 The charged particle beam system 100 according to the embodiments described herein may include a processing unit 170 and a memory 171, wherein a first correlation f1 between a first beam aberration coefficient and a first aberration characteristic may be stored in the memory 171, for example by storing scaling factors, value pairs ( X1 C1, ~ X1 C1) and / or ( X2 C1, ~ X2 Any one of C1) and / or any one or more of the y-intercepts of the determined linear function, such that the first correlation is fully defined. If the value of the first aberration characteristic retrieved from the actual defocused image is available, the first correlation f1 allows the determination of the actual value of the first bundle aberration coefficient.

[0075] In some embodiments, other correlations f between other bundle aberration coefficients and corresponding aberration characteristics are determined similarly to the first correlation f1. 1...n The values ​​of a set of actual bundle aberration coefficients can be determined and / or stored in memory 171 such that, if the values ​​of the aberration characteristics retrieved from the captured defocused image are known, the values ​​of such coefficients can be determined. Specifically, memory 171 can store multiple correlations between a set of bundle aberration coefficients and a set of corresponding aberration characteristics.

[0076] The set of beam aberration coefficients selected from the first beam aberration coefficient C1 may include any one or more of the following beam aberration coefficients: defocus (C df ), third-order spherical aberration (C s First-order astigmatism (A1), second-order astigmatism (A2), third-order astigmatism (A3), second-order coma (B2), third-order stellar aberration (S3), and / or one or more chromatic aberration coefficients.

[0077] In the following text, see references Figure 3 This describes how to determine the actual values ​​of the first beam aberration coefficients based on the first correlation f1. ac C1 method.

[0078] exist Figure 3 In block 310, one or more images of the sample are captured using one or more first defocus settings (step (d.)) to provide one or more captured images. The one or more first defocus settings for capturing one or more images in step (d.) may correspond to the one or more first defocus settings for simulating one or more beam cross-sections in step (a.). Specifically, the one or more first defocus settings may include one or more defocus distances (z) from the respective beam focus of the charged particle beam. o-1 , z u-1 This allows one or more images of the sample to be captured when the sample is positioned at one or more defocus distances from the beam focus, such as... Figure 1 The diagram is schematically illustrated. In an alternative approach, one or more images may be captured at one or more second defocus settings, which differ from one or more first defocus settings that simulate one or more beam cross-sections in step (a.). However, the latter may require interpolation based on at least two first correlations determined at the corresponding first defocus settings between the first beam aberration coefficients and the first aberration characteristics.

[0079] One or more defocus distances (z) are used to capture images in step (d.). o-1 , z u-1 ) can correspond to one or more defocus distances (z) of the multiple first simulated beam cross sections simulated in step (a.). o-1 , z u-1Specifically, one or more images can be obtained in step (d.) using numerical aperture, beam landing energy E. L And in step (a.), the image was taken under one or more first defocus settings simulating multiple first simulated beam cross sections.

[0080] In some embodiments that can be combined with other embodiments described herein, an overfocused image h of the sample can be captured. o-1 and underfocused image h of the sample u-1 Especially when the sample is positioned at the first overfocusing distance (z) from the beam focal point o-1 When the sample is positioned at the first underfocusing distance (z) from the beam focal point, and when the sample is positioned at the first underfocusing distance (z) from the beam focal point. u-1 (At that time) Figure 3 This schematically illustrates the first underfocusing distance (z) when the beam focus is located behind the sample. u-1 Underfocused image of the sample taken at (h) time u-1 ) and when the beam focus is positioned in front of the sample, the first overfocusing distance (z) o-1 Overfocus image of the sample taken at (h) o-1 Optionally, in some embodiments, a focused image h of the sample can also be captured. F This refers to a clear sample image taken at the focal point, for example, if a focused image of the sample cannot be obtained by other means.

[0081] The defocusing distance can be changed by altering the focusing intensity of the focusing lens 120, for example, by a predetermined increment (e.g., ...). Figure 1 (Illustrated schematically) The change. Specifically, an increased focusing intensity of the focusing lens 120 moves the corresponding beam focus and focusing plane relative to the sample toward the focusing lens, while a decreased focusing intensity moves the corresponding beam focus and focusing plane relative to the sample away from the focusing lens, thus changing the defocus distance. Alternatively or additionally, the defocus distance can be changed, for example, by moving the sample platform 108 in predetermined increments, particularly in the Z direction (along the optical axis A), while maintaining a constant focusing intensity provided by the focusing lens 120.

[0082] Alternatively or additionally, one or more images can be captured at one or more beam landing energies of charged particle beams that are different from the focusing beam landing energy, thereby capturing multiple images of the sample at one or more different beam landing energies. Similarly, here, one or more first defocus settings may correspond to one or more first defocus settings that simulate one or more beam cross-sections in step (a.), but the latter is not a necessary requirement, and one or more images can also be captured at one or more second defocus settings that are different from the one or more first defocus settings. As mentioned above, if the first beam aberration coefficient whose actual value is to be determined is a chromatic aberration beam aberration coefficient, it is particularly advantageous to simulate the beam cross-section at one or more different beam landing energies in step (a.) and capture defocused images at said one or more different beam landing energies in step (d.).

[0083] Variations in the beam landing energy of the focused beam (e.g., by changing the potential of the particle source (emitter) or the sample (wafer bias)) can shift the beam focus away from the sample, resulting in different beam landing energies producing a defocused image of the sample.

[0084] exist Figure 3 In frame 320, one or more retrieved beam cross sections ( retr g o-1 , retr g u-1 ) is from one or more captured images (h o-1 h u-1 The beam cross section retrieved is from one of one or more captured images, i.e., from each of the captured images. Therefore, if both underfocused and overfocused images of the sample are captured, the underfocus retrieved beam cross section ( retr g u-1 ) is retrieved from the underfocused image, and the beam cross section retrieved by overfocusing ( retr g o-1 () is retrieved from overfocused images.

[0085] In some embodiments that can be combined with other embodiments described herein, retrieving one or more retrieved beam cross sections from one or more images includes dividing one or more images in Fourier space by a focused image of the sample in Fourier space. Alternatively, from one or more images (h o-1 h u-1 Retrieve one or more retrieved beam cross sections from ) retr g o-1 , retr g u-1 This can be performed in real space, particularly based on deconvolution in real space, which is equivalent to division in Fourier space.

[0086] like Figure 3 As shown in box 320, from one or more captured images (h o-1 h u-1 Retrieve one or more retrieved beam cross sections from ) retr g o-1 , retr g u-1 Specifically, it can include the real space (h) o-1 h u-1 Perform a Fourier transform on one or more images captured, and convert one or more images (H) in Fourier space into their corresponding values. o-1 H u-1 Divide by the focused image of the sample in Fourier space (H F The described beam contour extraction method is based on the fact that, in Fourier space, the defocused image (H) of the sample is... o-1 Divide by the focused image of the sample (H) F The structure of the sample is removed by the division, such that the division produces a pure beam profile (G). o-1 This refers to the beam cross-section without sample information. Then, an inverse Fourier transform can be performed on the retrieved beam cross-sections in Fourier space to obtain one or more retrieved beam cross-sections in real space. retr g o-1 , retr g u-1 ).

[0087] From one or more captured images (h o-1 h u-1 Retrieve one or more retrieved beam cross sections from ) retr g o-1 , retr g u-1 Optionally, an applied filter may be included, particularly if the search is performed in Fourier space, in conjunction with an adaptive filter term. Multiplication, or convolution with an adaptive filter term if retrieval is performed in real space. The adaptive filter term can be provided by an adaptive filter unit that receives the captured image as input. The adaptive filter unit can provide an adaptive filter term individually for each of the captured images. Without an adaptive filter term, the focused image H in the denominator of the above division... F A value close to zero can lead to excessive noise weighting in the image. The adaptive filter term reduces or avoids focusing the image H when calculating one or more retrieved beam cross-sections. FThis undesirable effect of noise in the image can be mitigated, and a corresponding filter term can be individually determined for each captured image in the adaptive filter unit. As an alternative or supplement to the application of adaptive filter terms, retrieving one or more retrieved beam cross sections from one or more captured images can include focusing the beam cross section G in Fourier space. F Multiply.

[0088] exist Figure 3 In frame 330, from one or more retrieved bundle cross sections ( retr g o-1 , retr g u-1 Extract the retrieval value of the first aberration feature from (step (e.)). retr C1. Specifically, the retrieved value of the first aberration characteristic. retr C1 can be the beam cross-section retrieved from underfocusing. retr g u-1 and the beam cross section retrieved by overfocus retr g o-1 The retrieved value is obtained by using one of [Formula 1] and one of [Formula 3] above. retr C1, in particular, is similar to the determination of two or more values ​​of the first aberration characteristic in step (b.).

[0089] Specifically, it is possible to retrieve one or more beam cross sections ( retr g o-1 , retr g u-1 Each of the following defines one or more radial line profiles, and a retrieved value for a first aberration characteristic can be calculated based on the radial line profile. retr C1. Specifically, one or more of the following can be retrieved from each of one or more radial profiles: bundle width value, asymmetry value, and curvature value, and the retrieved value of the first aberration characteristic can be calculated accordingly. retr C1. Specifically, the retrieved value of the first aberration characteristic. retr C1 can be calculated from two or more values ​​of the first aberration characteristic from multiple first analog beam cross sections (see...). Figure 2 The corresponding method is used to retrieve one or more beam cross sections ( retr g o-1 , retr g u-1 The calculation is as described above and will not be repeated here.

[0090] exist Figure 3 In box 340, the retrieved value is based on the first relevance f1 and the first aberration characteristic. retrC1 determines the actual value of the first beam aberration coefficient (step (f)). ac C1, especially through ~ retr C1 is calculated or interpolated by the target value mapped to f1 (or by the inverse of f1, depending on how f1 is defined).

[0091] Since the calibration is performed using the absolute values ​​of the simulated first beam aberration coefficients, the actual values ​​of the first beam aberration coefficients determined based on the first correlation f1 are... ac C1 corresponds to the absolute value of the first beam magnification coefficient, for example, expressed in "mm".

[0092] In some embodiments, the aberration correction intensity of one or more beam aberration correctors may be based on the determined actual value of the first beam aberration coefficient used to correct the first beam aberration. ac Use C1 to set it.

[0093] In some embodiments that can be combined with other embodiments described herein, a second correlation between a second beam aberration coefficient and a second aberration characteristic in a set of beam aberration coefficients, and optionally, a further correlation between further beam aberration coefficients and corresponding further beam aberration characteristics, can be determined accordingly. In particular, a set of beam aberration coefficients C can be determined similarly to the first correlation f1. 1...n The second correlation f2 is the relationship between the second aberration coefficient C2 and the second aberration characteristic ~C2. The second correlation f2 can be based on different values ​​of the second aberration coefficient C2 at one or more first defocus settings or alternatively at one or more further defocus settings. x1 C2, x2 Multiple second simulated beam cross sections (g) of simulated defocusing of C2) o-2 g u-2 To determine. (See reference) Figure 2 And similar interpretations are applied to determine a second or further correlation. The second correlation and optional further correlations can be stored in the memory 171 of the charged particle beam system, so that the actual aberrations of the charged particle beam can be quickly determined based on the stored correlations.

[0094] Specifically, after capturing one or more images of the sample and retrieving one or more retrieved beam cross sections in step (d.), the retrieved values ​​of the second aberration characteristics can be extracted from the one or more retrieved beam cross sections in step (e.). retr C2 and / or further aberration characteristics retrieval values ​​~ retr C 1…n In step (f.), the retrieval value can be based on the second correlation f2 and the second aberration characteristic. retr C2 determines the actual value of the second beam aberration coefficient.ac C2, and / or in step (f.) may be based on further correlation f 1...n And based on further aberration characteristics~ retr C 1…n The retrieved value is used to determine further beam aberration coefficients. ac C 1...n The actual value.

[0095] In the above embodiments, the one or more first defocus settings simulating one or more beam cross sections in step (a.) correspond to the one or more first defocus settings for capturing one or more images in step (d.). In an alternative embodiment, the one or more defocus settings for capturing one or more images in step (d.) are one or more second defocus settings different from the one or more first defocus settings simulating one or more beam cross sections in step (a.). In the alternative embodiment, in step (a.), one or more beam cross sections may be simulated under two or more different first defocus settings, for example, under different pairs of underfocus and overfocus distances. In step (c.), at least two first correlations f between the first aberration coefficient C1 and the first aberration characteristic ~C1 are determined. 1-1 and f 1-2 Each of these two first correlations is based on the associated first defocus setting.

[0096] If at least two first correlation f are based on different first defocus settings 1-1 and f 1-2 If available, one or more images of the sample in step (d.) can be taken in one or more (arbitrary) second defocus settings that may differ from the first defocus setting used in the simulation. In step (f.), then based on at least two first correlation f... 1-1 and f 1-2 And the retrieval value is based on the first aberration characteristic. retr C1 is used to determine the actual value of the first aberration coefficient, specifically by using interpolation.

[0097] It should be noted that steps (a.) through (f.) do not necessarily have to be performed in the specified order. For example, before completing the simulation of one or more beam cross-sections (step (a.)), one or more images of the sample under one or more first defocus settings can be taken (step (d.)), and the first correlation f1 can be determined after taking one or more images. The latter order can provide the advantage that the simulation can be performed accurately at the first defocus setting of the actual captured images that can be obtained from the settings of the charged particle beam system.

[0098] Specifically, the correlation between (multiple) beam aberration coefficients and (multiple) corresponding aberration characteristics can be determined by simulation before, during, or after capturing actual images of the sample. However, the correlation f is determined before determining the actual aberrations of the charged particle beam from one or more defocused images (steps (d.)-(f.)). 1...n Steps (a.)-(c.)) may be beneficial because simulations can be time-consuming. In particular, it may be helpful if the correlation f is determined before actual measurements are performed. 1...n Furthermore, by storing this information in the memory of the charged particle beam system, it becomes feasible to quickly and reliably determine the actual beam aberration based on one or more captured defocused images.

[0099] According to one aspect described herein, a charged particle beam system is described. The charged particle beam system includes a simulation unit configured to perform the following operations: simulating one or more beam cross sections in one or more defocus settings for each of two or more distinct values ​​of a first beam aberration coefficient (C1) to provide a plurality of first simulated beam cross sections; extracting two or more values ​​of a first aberration characteristic from the plurality of first simulated beam cross sections; and determining a first correlation f1 between the first beam aberration coefficient (C1) and the first aberration characteristic (~C1) for determining the actual value of the first beam aberration coefficient. The first correlation f1 between the first beam aberration coefficient and the first aberration characteristic may optionally be stored in the memory of the charged particle beam system for later use when determining beam aberrations according to the methods described herein.

[0100] In some embodiments that can be combined with other embodiments described herein, the memory stores instructions that, when executed by a processing unit, cause the charged particle beam system to perform the following operations: retrieve one or more retrieved beam cross sections from one or more images captured in one or more defocus settings; extract retrieved values ​​of a first aberration characteristic from the one or more retrieved beam cross sections; and determine actual values ​​of a first beam aberration coefficient based on a first correlation and based on the retrieved values ​​of the first aberration characteristic. Actual values ​​of one or more further beam aberration coefficients can be determined similarly.

[0101] In some embodiments, the charged particle beam system further includes one or more aberration correctors, particularly one or more electrostatic or magnetic multipole correctors, which are configured to compensate for aberrations of the charged particle beam based on actual values ​​of a first beam aberration coefficient or actual values ​​of further beam aberration coefficients determined similarly.

[0102] The embodiments described herein provide a method for rapidly determining aberration coefficients in a charged particle beam system in absolute value form. The method is based on beam intensity distribution analysis of the beam cross-section used to determine aberration characteristics and calibration of these characteristics using wave optics simulation. The embodiments described herein achieve the speed of determination using a line profile-based method while allowing the determination of absolute values ​​of the beam aberration coefficients. The combination of rapid evaluation and absolute aberration values ​​is a unique feature of the method described herein. The method described herein can be combined with fitting routines for validity checks and / or refinement of the determined beam aberration coefficients. Furthermore, the determined absolute values ​​allow for performance comparisons between different tools.

[0103] In some embodiments, determined actual values ​​of the beam aberration coefficients(s) can be used to simulate the focused beam cross-section (i.e., focused beam spot shape) of an actual charged particle beam in a charged particle beam system. It is noteworthy that the focal spot shape of a charged particle beam cannot be easily measured directly, but the method described herein can simulate the focal spot shape with high accuracy. Beam spot shapes from different instruments can be retrieved and compared, for example, to improve the design of charged particle beam systems.

[0104] The analysis and determination of beam aberrations are prerequisites for the design or correction and reduction of harmful effects by aberration correctors, which may be part of the charged particle beam system described herein.

[0105] Specifically, the following embodiments are described herein:

[0106] Example 1: A method for determining the aberrations of a charged particle beam focused onto a sample by a focusing lens having a given numerical aperture (NA) in a charged particle beam system, the method comprising: (a.) based on a given numerical aperture (NA) and / or a given beam landing energy (E L For a set of beam aberration coefficients (C) 1...n Two or more distinct values ​​of the first aberration coefficient (C1) in ) x1 C1, x2 Each of C1) simulates one or more beam cross sections under one or more first defocus settings to provide a plurality of first simulated beam cross sections (g o-1 g u-1 (b.) Extract two or more values ​​(~C1) of the first aberration characteristic (~C1) associated with the first aberration coefficient (C1) from multiple first simulated beam cross sections. x1 C1, ~ x2 (c.) Determine a first correlation (f1) between the first aberration coefficient (C1) and the first aberration characteristic (~C1). Optionally, the first correlation may be stored in memory, for example, associated with a processing unit such as a computer.

[0107] The method may further include: (d.) capturing one or more images of the sample under one or more first defocus settings or one or more second defocus settings to provide one or more captured images (h o-1 h u-1 ), and retrieve one or more retrieved beam cross sections from one or more captured images ( retr g o-1 , retr g u-1 (e.) Extract the retrieved value of the first aberration characteristic (~C1) from one or more retrieved beam cross sections (~C1). retr (C1); and (f.) based on the first correlation and the retrieved value based on the first aberration characteristic, determine the actual value of the first bundle aberration coefficient.

[0108] Example 2: According to the method of Example 1, one or more first defocusing settings include one or more defocusing distances (z) from the focal point of the charged particle beam. o-1 , z u-1 ), such that in (a.), at one or more defocus distances (z) from the focal point. o-1 , z u-1 (a.) simulates one or more beam cross sections at (d.), and / or, when the sample is positioned at one or more defocus distances from the beam focus, captures one or more images of the sample. Specifically, in (a.), a pair of beam cross sections can be simulated for each of two or more distinct values ​​of the first beam aberration coefficient, i.e., at the overfocus distance (z) from the beam focus. o-1 The corresponding overfocusing cross section (g) at (i.e., on the first side of the beam focus) o-1 ), and the underfocusing distance (z) at the beam focal point. u-1 The corresponding underfocused beam cross section (g) at (i.e., on the second side of the beam focus) u-1 In (d.), when the sample is positioned at the overfocusing distance (z... o-1 When the sample is positioned at a distance of ) from the beam focal point, and when the sample is positioned at an underfocusing distance (z) from the beam focal point u-1 When at a certain location, one or more images of the sample can be captured.

[0109] In some embodiments, taking one or more images of the sample at one or more defocus distances in (d.) may include changing the focusing intensity of the focusing lens of the charged particle beam system, or alternatively changing the distance between the focusing lens and the sample, for example by moving the platform.

[0110] Example 3: The method according to Example 1 or 2, wherein (a.) includes simulating the focused beam cross section (g) for each of two or more different values ​​of the first beam aberration coefficient.o-1 ) and underfocused beam cross-section (g u-1 (b.) includes extracting each of two or more values ​​of the first aberration characteristic from one of the overfocused beam cross section and the corresponding underfocused beam cross section.

[0111] An overfocused beam cross-section can be simulated at an overfocusing distance (i.e., on the first side at a positive distance from the beam focus), and an underfocused beam cross-section can be simulated at a corresponding underfocusing distance (i.e., on the second side at a negative distance from the beam focus), where the absolute values ​​of the positive and negative distances can be the same.

[0112] Example 4: The method according to any one of Examples 1 to 3, wherein (d.) includes capturing an underfocused image and an overfocused image of the sample, and retrieving the underfocused beam cross section from the underfocused image and the overfocused beam cross section from the overfocused image, and / or (e.) includes extracting a retrieval value of a first aberration characteristic from the underfocused beam cross section and the overfocused beam cross section. Specifically, an underfocused image of the sample may be captured when the sample is arranged to be out of focus at an underfocused distance from the beam focus, and an overfocused image may be captured when the sample is arranged to be out of focus at an overfocused distance from the beam focus, wherein the absolute values ​​of the underfocused distance and the overfocused distance are the same but have different signs.

[0113] Example 5: The method according to any one of Examples 1 to 4, wherein in (b.), one or more radial line profiles are determined from each of the plurality of first simulated beam cross sections, and two or more values ​​of a first aberration characteristic are calculated based on the one or more radial line profiles. Specifically, the N radial line profiles can be obtained from each of the plurality of simulated beam cross sections surrounding the center of the respective simulated beam cross section at a constant angular step θ. k =kπ / N, k = 0...(N-1), where N can be a positive integer chosen based on the symmetry of the aberration coefficients to be analyzed.

[0114] Example 6: According to the method of Example 5, a bundle width value (σ), asymmetry value (μ), and / or curvature value (ρ) are retrieved from each of one or more radial line profiles, and two or more values ​​of the first aberration characteristic are calculated accordingly. Specifically, the bundle width value (σ), asymmetry value (μ), and / or curvature value (ρ) can be calculated based on the above [Formula 1].

[0115] In some embodiments, two or more values ​​of the first aberration characteristic (~ x1 C1, ~ x2 C1) can be determined based on one or the above [Formula 3], depending on what type of bundle aberration the first bundle aberration coefficient C1 is.

[0116] Example 7: According to the method described in Examples 1 to 6, a set of beam aberration coefficients (C 1...n This includes: defocusing (C) df ), third-order spherical aberration (C s First-order astigmatism (A1), second-order astigmatism (A2), third-order astigmatism (A3), second-order coma (B2), third-order stellar aberration (S3), and one or more chromatic aberrations (C). c The first beam aberration coefficient C1 can be any one of the beam aberration coefficients in the set of beam aberration coefficients.

[0117] Example 8: The method according to any one of Examples 1 to 7, wherein the first correlation between the first aberration coefficient (C1) and the first aberration characteristic (~C1) is a substantially linear correlation characterized by a first scaling factor, and (c.) includes determining the scaling factor.

[0118] Example 9: The method according to any one of Examples 1 to 8, wherein in (d.), retrieving one or more retrieved beam cross sections from one or more images comprises: dividing one or more images in Fourier space by a focused image of the sample in Fourier space. Optionally, retrieving one or more retrieved beam cross sections from one or more images may further comprise multiplying with an adaptive filter term and / or multiplying with at least one of the focused beam cross sections in Fourier space. In an alternative embodiment, one or more retrieved beam cross sections may be retrieved from one or more images based on deconvolution in real space.

[0119] Example 10: The method according to any one of Examples 1 to 9 further includes determining a set of bundle aberration coefficients (C 1...n A second correlation is established between the second beam aberration coefficient (C2) and the second aberration characteristic (~C2) in the model, specifically, wherein the second correlation is determined based on multiple second simulated beam cross sections simulated for different values ​​of the second beam aberration coefficient (C2). Optionally, one or more further correlations may be determined between one or more further beam aberration coefficients and one or more further aberration characteristics.

[0120] Example 11: The method according to Example 10 further includes: extracting a retrieval value of a second aberration characteristic from one or more retrieved beam cross sections; and optionally extracting a retrieval value of at least one further aberration characteristic from one or more retrieved beam cross sections; and determining an actual value of a second beam aberration coefficient based on a second correlation and the retrieval value of the second aberration characteristic, and optionally determining an actual value of at least one further beam aberration coefficient based on at least one further correlation and the retrieval value of at least one further aberration characteristic.

[0121] Example 12: The method according to any one of Examples 1 to 11 further includes storing a first correlation in the memory of the charged particle beam system, wherein the first correlation is a substantially linear correlation characterized by a first scaling factor stored in the memory. Further correlations between further beam aberration coefficients in a set of beam aberration coefficients and their corresponding aberration characteristics can be determined similarly to the first correlation in steps (a.)-(c.) and may also optionally be stored in the memory.

[0122] Example 13: The method according to any one of Examples 1 to 12, wherein in (c.), at least two first correlations between a first beam aberration coefficient (C1) and a first aberration characteristic are determined, each first correlation being based on an associated first defocus setting, and in particular, each first correlation being based on a corresponding pair of overfocus beam cross sections and underfocus beam cross sections; in (d.), one or more images of the sample are captured under one or more second defocus settings different from one or more first defocus settings; and in (f.), the actual value of the first beam aberration coefficient is determined, in particular, using interpolation, based on at least two first correlations and based on a retrieved value of the first aberration characteristic.

[0123] Example 14: The method according to any one of Examples 1 to 13, wherein one or more first defocus settings include one or more beam landing energies of charged particle beams varying from the focusing beam landing energy, such that in (a.), defocus simulation is performed on one or more beam cross sections at one or more beam landing energies, and in (d.), one or more images of the sample are defocused at one or more beam landing energies. In some embodiments, the one or more beam cross sections are defocus simulated at one or more different defocus distances, and the one or more beam cross sections are defocus simulated at one or more different beam landing energies. In addition to the correlation between “geometric” aberration coefficients and corresponding aberration characteristics, the correlation between “chromatic” aberration coefficients and corresponding aberration characteristics can also be determined.

[0124] Example 15: The method according to any one of Examples 1 to 14 further includes setting the aberration correction intensity of one or more beam aberration correctors based on the actual value of the first beam aberration coefficient and optionally based on a further actual value of the beam aberration coefficient determined similarly to the actual value of the first beam aberration coefficient.

[0125] Example 16: A method for determining the aberrations of a charged particle beam focused onto a sample by a focusing lens having a given numerical aperture (NA) in a charged particle beam system, the method comprising: at least based on the given numerical aperture (NA), for a set of beam aberration coefficients (C 1...nEach of two or more distinct values ​​of the first beam aberration coefficient (C1) in the first beam aberration coefficient (C1) is used to simulate one or more beam cross sections under one or more first defocus settings to provide a plurality of first simulated beam cross sections; two or more values ​​of a first aberration characteristic associated with the first beam aberration coefficient (C1) are extracted from the plurality of first simulated beam cross sections; a first correlation between the first beam aberration coefficient (C1) and the first aberration characteristic is determined; and the first correlation is stored in memory for subsequent determination of the actual value of the first beam aberration coefficient. Embodiment 16 may be combined with or may additionally include features of any of Embodiments 1 to 15 described above.

[0126] Example 17: A charged particle beam system, particularly a charged particle beam system configured for performing any of the methods described herein, the charged particle beam system comprising: a charged particle source for emitting a charged particle beam propagating along an optical axis; a sample platform; a focusing lens for focusing the charged particle beam having a given numerical aperture (NA) onto a sample placed on the sample platform; a charged particle detector for detecting signal particles emitted from the sample; and a processing unit and a memory. The memory stores a first correlation between a first beam aberration coefficient (C1) and a first aberration characteristic (~C1), the first aberration characteristic being a parameter having a predetermined relationship with the first beam aberration coefficient and extractable from one or more defocused beam cross sections.

[0127] Example 18: According to the charged particle beam system of Example 17, wherein the memory stores instructions that, when executed by the processing unit, cause the charged particle beam system to perform the following operations: retrieve one or more retrieved beam cross sections from one or more images captured in one or more defocus settings; extract a retrieval value of a first aberration characteristic from the one or more retrieved beam cross sections; and determine an actual value of a first beam aberration coefficient based on a first correlation and based on the retrieval value of the first aberration characteristic.

[0128] Example 19: The charged particle beam system according to Example 17 or 18 further includes one or more aberration correctors, particularly one or more electrostatic or magnetic multipole correctors, said one or more aberration correctors being configured to compensate for aberrations of the charged particle beam based on the actual value of the aberration coefficient of the first beam.

[0129] Example 20: A charged particle beam system according to any one of Examples 17 to 19, wherein the memory stores multiple correlations between a set of beam aberration coefficients and a set of corresponding aberration characteristics.

[0130] Example 21: A charged particle beam system according to any one of Examples 18 to 20, wherein one or more defocus settings include one or more defocus distances, particularly underfocus distances and corresponding overfocus distances, of the sample from the corresponding focal point of the charged particle beam when one or more images are captured.

[0131] Example 22: A charged particle beam system according to any one of Examples 18 to 21, wherein when capturing one or more images, one or more first defocus settings include one or more beam landing energies of charged particle beams varying from the landing energy of the focusing beam.

[0132] Example 23: A charged particle beam system, particularly according to any one of Examples 17 to 22, comprising a simulation unit configured to perform the following operations: simulating one or more beam cross sections in one or more defocus settings for each of two or more distinct values ​​of a first beam aberration coefficient (C1) to provide a plurality of first simulated beam cross sections; extracting two or more values ​​of a first aberration characteristic from the plurality of first simulated beam cross sections; and determining a first correlation between the first beam aberration coefficient (C1) and the first aberration characteristic (~C1) to determine the actual value of the first beam aberration coefficient. The charged particle beam system may further include any of the features of the embodiments described herein and / or may be configured to perform any of the methods described herein.

[0133] Although the foregoing describes an embodiment, other and further embodiments may be devised without departing from its basic scope, the scope of which is defined by the appended claims.

Claims

1. A method for determining the aberrations of a charged particle beam focused onto a sample by a focusing lens having a given numerical aperture in a charged particle beam system, the method comprising: (a.) At least based on the given numerical aperture, for each of two or more distinct values ​​of a first beam aberration coefficient in a set of beam aberration coefficients, simulate one or more beam cross sections under one or more first defocus settings to provide a plurality of first simulated beam cross sections; (b.) Extract two or more values ​​of a first aberration characteristic associated with the aberration coefficients of the first beam from the plurality of first simulated beam cross sections; (c.) Determine a first correlation between the first beam aberration coefficients and the first aberration characteristic; (d.) To capture one or more images of the sample under one or more first defocus settings or one or more second defocus settings to provide one or more captured images, and to retrieve one or more retrieved beam cross sections from the one or more captured images; (e.) Extract the retrieved value of the first aberration characteristic from the one or more retrieved beam cross sections; as well as (f.) Based on the first correlation and the retrieved value based on the first aberration characteristic, determine the actual value of the first bundle aberration coefficient.

2. The method of claim 1, wherein the one or more first defocus settings include one or more defocus distances from the focal point of the charged particle beam, such that in (a.), the one or more beam cross-sections are simulated at the one or more defocus distances from the focal point, and in (d.), the one or more images of the sample are captured when the sample is positioned at the one or more defocus distances from the focal point.

3. The method of claim 1, wherein (a.) includes simulating an overfocused beam cross section and an underfocused beam cross section for each of the two or more distinct values ​​of the first beam aberration coefficient, and (b.) includes extracting each of the two or more values ​​of the first aberration characteristic from one of the overfocused beam cross section and the corresponding underfocused beam cross section.

4. The method of claim 1, wherein (d.) comprises capturing an underfocused image of the sample and an overfocused image of the sample, and retrieving a beam cross section retrieved from the underfocused image and a beam cross section retrieved from the overfocused image, and (e.) comprises extracting the retrieved value of the first aberration characteristic from the beam cross section retrieved from the underfocused image and the beam cross section retrieved from the overfocused image.

5. The method of claim 1, wherein in (b.), one or more radial line profiles are determined from each of the plurality of first simulated beam cross sections, and the two or more values ​​of the first aberration characteristic are calculated based on the radial line profiles.

6. The method of claim 5, wherein one or more of a bundle width value, asymmetry value, and curvature value are retrieved from each of the one or more radial profiles, and the two or more values ​​of the first aberration characteristic are calculated thereon.

7. The method of claim 1, wherein the first beam aberration coefficient is selected from the set of beam aberration coefficients, the set of beam aberration coefficients including defocus (C df ), third-order spherical aberration (C s First-order astigmatism (A1), second-order astigmatism (A2), third-order astigmatism (A3), second-order coma (B2), third-order stellar aberration (S3), and one or more chromatic aberrations (C). c ).

8. The method of claim 1, wherein the first correlation between the first aberration coefficient and the first aberration characteristic is a substantially linear correlation characterized by a first scaling factor, and (c.) includes determining the first scaling factor.

9. The method of claim 1, wherein in (d.), retrieving the one or more retrieved beam cross sections from the one or more images comprises: Divide the one or more images in Fourier space by the focused image of the sample in Fourier space, or alternatively, retrieve the one or more retrieved beam cross sections based on deconvolution in real space.

10. The method of claim 1, further comprising determining a second correlation between a second beam aberration coefficient and a second aberration characteristic in the set of beam aberration coefficients.

11. The method of claim 10, wherein the second correlation is determined based on a plurality of second simulated beam cross sections simulated for different values ​​of the second beam aberration coefficient.

12. The method of claim 10, further comprising: Extract the retrieved value of the second aberration characteristic from the one or more retrieved beam cross sections; as well as Based on the second correlation and the retrieved value based on the second aberration characteristic, the actual value of the second bundle aberration coefficient is determined.

13. The method of claim 1, further comprising storing the first correlation in the memory of the charged particle beam system.

14. The method of claim 13, wherein the first correlation is a substantially linear correlation characterized by a first scaling factor stored in the memory.

15. The method of claim 1, wherein in (c.), at least two first correlations between the first bundle aberration coefficients and the first aberration characteristic are determined, each first correlation being based on an associated first defocus setting; In (d.), the one or more images of the sample were taken under one or more second defocus settings that are different from the one or more first defocus settings; as well as In (f.), the actual value of the first aberration coefficient is determined based on the at least two first correlations and the retrieved value based on the first aberration characteristic.

16. The method of claim 1, wherein the one or more first defocus settings include one or more beam landing energies of the charged particle beam varying from the focused beam landing energy, such that in (a.), the cross-section of the one or more beams is defocused at the one or more beam landing energies, and in (d.), the one or more images of the sample are defocused at the one or more beam landing energies.

17. The method of claim 1, further comprising setting the aberration correction intensity of one or more beam aberration correctors based on the actual value of the first beam aberration coefficient.

18. A method for determining the aberrations of a charged particle beam focused onto a sample by a focusing lens having a given numerical aperture in a charged particle beam system, the method comprising: (a.) At least based on the given numerical aperture, for each of two or more distinct values ​​of the first beam aberration coefficient, simulate one or more beam cross sections under one or more first defocus settings to provide a plurality of first simulated beam cross sections; (b.) Extract two or more values ​​of a first aberration characteristic associated with the aberration coefficients of the first beam from the plurality of first simulated beam cross sections; as well as (c.) Determine a first correlation between the first beam aberration coefficient and the first aberration characteristic to determine the actual value of the first beam aberration coefficient.

19. A charged particle beam system, comprising: A charged particle source, the charged particle source being used to emit a beam of charged particles propagating along the optical axis; Sample platform; A focusing lens for focusing a beam of charged particles having a given numerical aperture onto a sample placed on the sample platform; A charged particle detector for detecting signal particles emitted from the sample; as well as The processing unit and the memory store a first correlation between first beam aberration coefficients and a first aberration characteristic, the first aberration characteristic being a parameter that has a predetermined relationship with the first beam aberration coefficients and can be extracted from one or more defocused beam cross sections. The memory stores instructions that, when executed by the processing unit, cause the charged particle beam system to perform the following operations: Retrieve one or more retrieved beam cross sections from one or more images captured in one or more defocus settings; Extract the retrieved value of the first aberration characteristic from the one or more retrieved beam cross sections; as well as Based on the first correlation and the retrieved value based on the first aberration characteristic, the actual value of the first bundle aberration coefficient is determined.

20. The charged particle beam system of claim 19, further comprising one or more aberration correctors configured to compensate for aberrations of the charged particle beam based on the actual values ​​of the aberration coefficients of the first beam.

21. The charged particle beam system of claim 20, wherein the aberration corrector comprises an electrostatic multipole corrector or a magnetic multipole corrector.

22. The charged particle beam system of claim 19, wherein the memory stores multiple correlations between a set of beam aberration coefficients and a set of corresponding aberration characteristics.

23. The charged particle beam system of claim 19, wherein the charged particle beam system includes a simulation unit configured to perform the following operations: For each of two or more distinct values ​​of the first beam aberration coefficient, one or more beam cross sections are simulated under one or more defocus settings to provide multiple first simulated beam cross sections; Extract two or more values ​​of the first aberration characteristic from the plurality of first simulated beam cross sections; as well as The first correlation between the first beam aberration coefficient and the first aberration characteristic is determined to determine the actual value of the first beam aberration coefficient.

Images