WO2015037313A1

WO2015037313A1 - Scanning transmission electron microscope and aberration measurement method therefor

Info

Publication number: WO2015037313A1
Application number: PCT/JP2014/068078
Authority: WO
Inventors: 高穂吉田
Original assignee: 株式会社日立ハイテクノロジーズ
Priority date: 2013-09-13
Filing date: 2014-07-07
Publication date: 2015-03-19
Also published as: JP2015056376A; JP6163063B2

Abstract

Acquisition of aberration information from a ronchigram is prone to occurrences of errors. Thus, when imaging a ronchigram, the positional relationship between the focal point of an electron probe imaged onto a sample surface or in the vicinity thereof is varied minutely in the height direction, and the acquired ronchigram (focus-modulated ronchigram) is used to measure the aberration.

Description

Scanning transmission electron microscope and aberration measurement method thereof

The present invention relates to a technique for precisely measuring the aberration of a scanning transmission electron microscope (STEM).

In recent years, spherical aberration correction technology has been put into practical use in scanning transmission electron microscopes (STEM) and transmission electron microscopes (TEM). Along with this, a method of appropriately handling the influence of various minute aberrations that have been hidden behind the large spherical aberration of the electron lens and have not been emphasized so far has become important. For example, before the spherical aberration corrector was put to practical use, the third-order and lower aberrations that are problematic in STEM are at most three aberrations: defocus, two-fold astigmatism, and three-fold symmetry astigmatism. Or, it would have been good to take compensation measures.

However, since the spherical aberration corrector was put into practical use, in addition to the three aberrations, the remaining four aberrations after correcting the spherical aberration (axial coma, star aberration, four-fold astigmatism, and spherical aberration itself) It is necessary to handle a total of 7 aberrations including In order to obtain the highest resolution, it is necessary to evaluate and adjust or suppress a total of 14 axial aperture aberrations of higher order (generally 5th order or less in 2012).

As is clear from the above, the evaluation of each lens aberration (that is, the precision measurement method of the aberration coefficient) is becoming more important than before the establishment of the electron microscope aberration correction technology. Conventionally, in an aberration correction STEM device (STEM device including a spherical aberration corrector), the probe tableau method (Non-patent Document 1) has been mainly used as a method for measuring precise aberrations during aberration correction. In this method, a plurality of STEM images with appropriate defocusing are given by varying the incident angle of the electron beam (by a plurality of incident angles). This is a method of performing deconvolution calculation using a sufficiently small STEM image) as a reference image, estimating distortion of the electron probe due to aberration at each electron beam incident angle, and calculating axial aberration from these estimation results.

From the estimated probe shape, the defocus C ₁ (τ) and the two-fold astigmatism A ₁ (τ) at the incident angle at which the original STEM image was taken can be obtained. Are measured, a multiple simultaneous equation is established from the relational expression of axial aberration and tilt aberration, which will be described later, and as a result, the coefficient of each axial aberration listed in Table 1 is calculated. Here, τ attached to each aberration coefficient indicates designation of the incident angle of the electron beam. Among the aberrations shown in Table 1, non-rotationally symmetric aberrations (that is, aberrations with symmetry in 1 to 6 rows) have two parameters, magnitude and direction. 25 unconstant must be determined. That is, it is necessary to establish at least as many independent simultaneous equations as there are undetermined numbers.

In the probe tableau method, it is possible to obtain a total of three equations of C ₁ (τ) magnitude, A ₁ (τ) magnitude and orientation with respect to the incident angle of one electron beam. If these can be measured using angles, axial aberrations up to the fifth order can be obtained. In an actual example, in order to statistically improve the measurement accuracy, S ₁ T images were taken with appropriate underfocus and overfocus at 17 to 25 incident angles, and C ₁ (τ) and A ₁ (τ) were calculated. And calculate the axial aberration.

That is, adding one image at the normal incidence normal focal point as a reference image and taking 35 or more STEM images to execute the aberration measurement makes measurement extremely complicated and time consuming. Such an aberration measurement method has a problem that the calculated on-axis aberration coefficient is likely to contain errors not only due to errors occurring in the process of estimating the STEM image itself or the shape of the electron probe but also due to changes over time during STEM image shooting. is there.

On the other hand, in recent years, an aberration measurement method using a projected image (Ronchigram) (for example, Patent Document 1 and Patent Document 2) has been developed and put to practical use as an aberration measurement method based on a principle different from the probe tableau method. Although details of the Ronchigram will be described later, the Ronchigram itself is a projected figure of the sample by the electron probe distorted by the aberration, and the figure includes information on the aberration.

Furthermore, if a Ronchigram is imaged using an electron beam having a sufficiently large convergence angle, aberration information relating to the incident electron beam within the convergence angle is projected onto the corresponding location of the Ronchigram. For this reason, unlike the above-described probe tableau method, it is possible in principle to calculate axial aberration from one Ronchigram by extracting these aberration information for each local area of the Ronchigram. The axial aberration can be calculated from the Ronchigram of several conditions including the repetition for the purpose of statistically improving the measurement accuracy. However, the Ronchigram including aberration becomes an atypical image having complicated distortion. Therefore, in order to extract aberration information from an atypical image, appropriate image processing and calculation are essential.

For example, in Patent Document 1, the obtained Ronchigram is divided into an appropriate size and the required number of lattices, the image distortion in each lattice and the autocorrelation of the lattice image are taken, and the pattern obtained by the autocorrelation is elliptical. Quantify by fitting with a function. The parameters that determine the elliptic function obtained by fitting are linked to the local aberration information in the lattice using Equation 3, Equation 4, and Equation 5 in the same document, and further, a simultaneous equation obtained from Equation 2 is established from each lattice. The axial aberration coefficient is calculated.

In this case, of course, the number of gratings is sufficient to establish the simultaneous equations necessary for the calculation of the on-axis aberration, but the size of each grating must also be an appropriate size to extract the local aberration information of the area. I must. Since what is obtained from the autocorrelation of each region divided by the lattice is the strain information averaged within the region, it is preferable that the size of the lattice is as small as possible to know the strain information corresponding to the local area. On the other hand, if the lattice size is too small, the autocorrelation correctly reflecting the distortion information cannot be obtained, and an error occurs in the obtained local aberration.

By the way, the shape of the Ronchigram varies greatly depending on the state of aberration. For this reason, the size of the grating needs to be appropriately selected according to the aberration state. For defocus and camera length (Ronchigram magnification) at the time of Ronchigram acquisition, careful selection according to the aberration state is necessary for the same reason.

Patent Document 2 and Patent Document 3 also disclose an aberration measurement method using a Ronchigram. These documents disclose methods for measuring aberrations by measuring changes in the Ronchigram with varying optical parameters.

Among these, Patent Document 2 photographs a Ronchigram before and after the electron probe focused in the vicinity of the sample surface is slightly displaced by an electron optical means such as a deflector (or the sample itself is minutely displaced). The axial aberration is measured by comparing these two Ronchigrams. Specifically, as in Patent Document 1, each Ronchigram is appropriately divided into a plurality of regions, and the local displacement generated before and after the minute displacement of the electron probe is measured in a corresponding region between the two Ronchigrams. Since the local displacement in each region can be regarded as a change in local magnification due to aberration, use Equation 1 in the same document that relates the local displacement and axial aberration, and also calculate the axial aberration by combining simultaneous equations from each region. To do.

Further, in Patent Document 3, instead of slightly displacing the electron probe, the focal point of the imaging lens is slightly changed, and Ronchigrams are obtained before and after the electron probe is moved up and down in the vicinity of the sample, and the two Ronchigrams are compared. As a result, the axial aberration is measured. Displacement is measured between a plurality of corresponding projection points of two Ronchigrams acquired before and after defocusing is slightly changed, and this is correlated with axial aberration as local aberration information. To calculate the on-axis aberration.

In the methods shown in

Patent Documents

2 and 3, it is necessary to note that the extraction of local aberration information is appropriate (and the method of dividing the Ronchigram is also divided) as in the method of Patent Document 1. Furthermore, unlike the method of Patent Document 1, since the displacements of the corresponding projection points are compared between two different Ronchigrams, it is necessary to correctly associate the positions before and after the displacement at a plurality of projection points. On the other hand, in the methods of Patent Document 2 and Patent Document 3, unlike the method of Patent Document 1, the amount of displacement in the Ronchigram can be adjusted by the amount of displacement and defocus of the electronic probe that can be arbitrarily controlled externally. Thus, it is possible to improve the extraction accuracy of the local aberration information and hence the measurement accuracy of the on-axis aberration.

Japanese Patent No. 5188846 US Pat. No. 6,552,340 Japanese Patent No. 4553889

As described above, the probe tableau method that has been mainly used in STEM conventionally requires a complicated and time-consuming procedure for repeating many measurements while changing the incident angle of the electron beam. It is difficult to say that it is an easy-to-use measurement method for use in the correction (adjustment) process.

On the other hand, the aberration measurement method using the Ronchigram can measure the aberration coefficient that gives the axial aberration by a smaller number of measurements (in a short time) than the probe tableau method. However, the aberration measurement methods using these Ronchigrams also have a number of problems in the image measurement process for extracting a large amount of local aberration information from the Ronchigram as described above.

In order to solve the above-described problems, the present invention minutely changes the height relationship between the focal point of the electron probe imaged on or near the sample surface and the sample when imaging one Ronchigram. Get Ronchigram while letting In the present specification, the Ronchigram acquired while slightly changing the focal point as described above is referred to as “Focus Modulated Ronchigram”.

According to the present invention, local aberration information can be extracted with only one Ronchigram (focus modulation Ronchigram). In addition, in the case of the present invention, aberration information can be accurately extracted from the Ronchigram as compared with the conventional method, and as a result, it is possible to accurately measure even higher-order aberrations. Problems, configurations, and effects other than those described above will become apparent from the following description of embodiments.

Schematic for demonstrating the structural example of STEM provided with a spherical aberration corrector. The figure explaining the formation principle of Ronchigram. The figure explaining the influence which a focus change has on a Ronchigram (especially the displacement of the projection point on the Ronchigram corresponding to the specific point on a sample). 1 is a diagram illustrating a configuration example of an STEM according to Example 1. FIG. FIG. 6 is a diagram illustrating a configuration example of an STEM according to the second embodiment. FIG. 10 is a diagram illustrating a configuration example of an STEM according to the third embodiment. The figure which shows the example of a photograph of an amorphous thin film sample. The figure which shows the normal Ronchigram obtained by imaging the amorphous thin film sample shown to FIG. 7A. FIG. 7B is a diagram showing a focus modulation Ronchigram obtained by imaging the amorphous thin film sample shown in FIG. 7A. The figure explaining the process of taking out a local region from a focus modulation Ronchigram. The figure explaining the process of dividing | segmenting the taken-out area | region into suitable size. The figure which shows the result of having calculated | required the autocorrelation about each area | region after a division | segmentation. The figure explaining the local aberration information extracted from the focus modulation Ronchigram. The figure explaining the method (especially the method of acquiring the line element showing elongation distortion from correlation intensity | strength) which digitizes local aberration information.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. The embodiment of the present invention is not limited to the examples described later, and various modifications are possible within the scope of the technical idea.

This specification provides a method for measuring axial aberration with high accuracy in STEM, particularly an aberration measurement method suitable for adjustment of an aberration correction STEM apparatus. In particular, the present specification relates to an aberration measurement method using a Ronchigram, and provides a high-accuracy aberration coefficient measurement method that calculates a quantitatively accurate on-axis aberration coefficient by reducing the arbitraryness at the time of measurement.

[Basic configuration]
First, the basic configuration of the aberration correction STEM apparatus will be described. FIG. 1 shows a schematic configuration of an aberration correction STEM apparatus. The aberration correction STEM device includes an electron source 3 and acceleration means (for example, an acceleration tube 2) for accelerating electrons emitted from the electron source 3 at the top of the mirror body 1. In addition, there is an apparatus provided with the electron gun 2 at the lowermost end as in some STEMs (VG STEM apparatus and the like). In this case, the apparatus configuration is opposite to that shown in FIG. However, whether the electron source 2 is at the top or the bottom as shown in FIG. 1 does not have a particular influence on the present invention.

The electron beam 20 is emitted from the electron source 3. The trajectory of the electron beam 20 that has passed through the converging lens 4 is adjusted by the deflector 6, and the converging state is adjusted by the converging lens 7. These adjusted electron beams 20 are guided to the objective lens 10. The objective lens 10 converges the electron beam 20 on the surface of the sample 9 as an electron probe having a minute diameter. The electronic probe is raster scanned by the scan coil 24 along the sample surface. At this time, transmitted electrons, scattered electrons, secondary electrons, reflected electrons, X-rays, and the like are obtained from various portions of the sample 9.

Among these, the transmission electron beam 22 and the scattered electron beam 23 are adjusted by the projection lens 12 so as to have an appropriate detection angle with respect to the bright field detector 14 and the annular dark field detector 13. Thereby, the transmission electron beam 22 is detected by the bright field detector 14, and the scattered electron beam 23 is detected by the annular dark field detector 13.

These detection signals are read out to the control computer (control PC) 19 through the electron microscope control system 17. The control computer 19 reconstructs the detection signal into a two-dimensional image in the order of raster scanning on the sample surface. As a result, a bright field STEM image is obtained from the detection signal of the bright field detector 14 (corresponding to the transmission electron beam 22), and the annular dark field is detected from the detection signal of the dark field detector 13 (corresponding to the scattered electron beam 23). A STEM image is obtained. As can be inferred from the above description, in order to obtain a high resolution in the STEM apparatus, it is important how to form an electron probe with a minute diameter size on the surface of the sample 9 in the objective lens 10.

Conventionally, it has been very difficult to reduce the diameter of an electron probe to 0.1 nm or less due to the influence of positive spherical aberration that cannot be avoided with an objective lens that is an electromagnetic field lens. However, due to the practical application of the spherical aberration corrector 11 in recent years, the spherical aberration of the objective lens 10 can be corrected, and the diameter of the electron probe can be reduced to 0.1 nm or less. In fact, in the aberration correction STEM apparatus, the diameter of the electron probe can be reduced to less than 0.1 nm (that is, 0.08 to 0.05 nm), and it has become possible to obtain a resolution that can clearly identify individual atoms.

In such an aberration correction STEM device, the spherical aberration corrector 11 generates a negative (opposite) spherical aberration equivalent to the spherical aberration of the objective lens 10 in advance when forming the electron probe. As shown, it is provided above the objective lens 10. The spherical aberration corrector 11 is not a rotationally symmetric electromagnetic lens like a conventional electron lens, but has a structure in which a plurality of multipole lenses such as hexapoles, quadrupoles, octupoles, etc. are used in combination. For this reason, the adjustment of the spherical aberration corrector 11 is complicated, and an aberration correction control system 18 for controlling the adjustment is required. In an actual hardware configuration, the aberration correction control system 18 may be integrated into the electron microscope control system 17 and the control computer 19. In any case, the aberration correction control system 18 is a new apparatus or software that has not been used in a conventional electron microscope (STEM or the like).

The aberration correction control system 18 includes a power source that drives the main body of the spherical aberration corrector 11, an aberration measuring unit that evaluates the aberration state of the STEM, and an adjustment unit that performs correction adjustment according to the evaluated aberration state. The By the way, the configuration and control of the spherical aberration corrector 11 are complicated, and the adjustment needs to be performed precisely. Therefore, in the normal aberration correction procedure, 1) aberration measurement is executed first, 2) the adjustment content is determined with reference to the measurement result, 3) adjustment is executed, and 4) aberration measurement is executed again. 5) Confirming the effect of adjusting the aberration, 6) Cyclicly repeating a series of processes such as determining the next adjustment content, and asymptotically leading the spherical aberration corrector 11 and the STEM apparatus to the desired aberration correction state. Adjustment operation is adopted.

However, such adjustment operations involve complicated aberration measurement and aberration adjustment itself, and it takes a very long time to complete the adjustment, or in the worst case, the STEM device is in the middle of adjustment. There is a possibility that the aberration adjustment cannot be properly completed due to an influence of thermal drift or the like.

[Principle of Aberration Measurement Method Used in Examples]
In examples described later, Ronchigram is used for aberration measurement. For this reason, the projection surface 15 and the imaging detector 16 are disposed below the mirror body. The Ronchigram is an image (projected image) of the sample 9 projected onto the projection surface 15 when the beam scanning by the scan coil 24 is stopped.

When acquiring Ronchigram, use the converging lens aperture 5 with a large aperture diameter. In this case, the electron beam 20 transmitted through a wide range of the objective lens 10 is converged to the vicinity of the sample 9 by the objective lens 10. At this time, the electron probe has a distorted shape due to the influence of aberration at various points of the objective lens 10. A Ronchigram (projected image) of the sample 9 formed by the distorted electron probe is projected onto the projection surface 15 via the projection lens 12. The Ronchigram is photographed by an imaging detector 16 such as a CCD camera that photographs the projection surface 15 from the back surface side, and is taken into the control computer 19 as image data.

The principle of Ronchigram formation will be described with reference to FIG. The electron beam 20 that has passed through the spherical aberration corrector 11 enters the objective lens 10. At this time, when the objective lens 10 is ideal (has no aberration), the electron beam 20 passes through the

trajectories

21a and 210a and is converged to one point (convergence point O) on the sample surface. In FIG. 2, the ideal wavefront of the electron beam 20 with respect to convergence in the ideal state is represented by 30a. As shown in the figure, the ideal wavefront 30a is a spherical surface centered on the convergence point O.

In this case, in the Ronchigram projected on the projection surface 150, the sample 9 at the convergence point O is evenly (uniformly) projected onto all regions in the Ronchigram. Therefore, if there is a particle that blocks the electron beam 20 at the convergence point O, the entire surface of the Ronchigram is uniformly darkened. On the other hand, if there is a hole that transmits the electron beam 20, the entire surface of the Ronchigram becomes uniformly bright and no conspicuous pattern appears. This is a Ronchigram with no aberration. Therefore, the state to be approximated using the spherical aberration corrector 11 is a Ronchigram having a uniform region having no characteristic pattern as wide as possible.

On the other hand, when the objective lens 10 has aberration, the electron beam 20 is not converged to one point (convergence point O) on the sample surface as in an ideal state. For example, the electron beam 210b passing through the point Q of the objective lens 10 passes through a point P that is separated from the convergence point O on the sample surface by δ due to the influence of aberration, passes through the trajectory of the electron beam 221b, and is a point in the Ronchigram. R is reached. That is, the Ronchigram affected by the aberration becomes a distortion projection image of the sample 9. When the aberration is included, the wavefront is also deformed from the ideal wavefront 30a to the wavefront 30b including the aberration. This wavefront difference is the wavefront aberration. The wavefront aberration (the left side of Equation 1) is expressed by the following equation using the aberration count.
[Formula 1]

Equation 1 specifies up to fifth-order aberrations. In the equation, ω is a complex representation of the electron convergence angle and is given by Equation 2, respectively.
[Formula 2]

Here, α and β are incident angles with respect to in-plane and perpendicular to the plane, respectively. Therefore, in FIG. 2, only α is shown. Displacement δ on the sample surface due to aberration is expressed by the following equation by complex display using α and β.
[Formula 3]

Further, the projection point R in the Ronchigram is given by the following equation.
[Formula 4]

In Expression 4, L is a so-called camera length (distance from the sample 9 to the image plane 150). As expressed in Expression 4, the projection point R on the Ronchigram corresponding to a certain point of the sample 9 is determined by the influence of wavefront aberration.

Next, how the projected point R in the Ronchigram changes when the electron probe is slightly defocused in the vicinity of the sample 9 from the state of FIG. 2 will be described with reference to FIG. Defocusing means that an ideal focal plane (Gaussian focal point) having no aberration moves from the convergence point 0 to the convergence point E. At this time, the trajectory (paraxial trajectory) of the electron beam 20 due to ideal (without aberration) convergence changes from the trajectory 210a to the trajectory 211a. The ideal wavefront changes from the wavefront 30a to the wavefront 31a, and the wavefront including aberration changes from the wavefront 30b to 31b.

When defocusing occurs, as shown in FIG. 3, the electron beam 20 passing through the trajectory 210b from the point Q of the objective lens 10 no longer passes through the point P on the sample 9, and starts from a new point Q ′ of the objective lens 10. The electron beam 20 passing through the orbit 211b irradiates the point P. That is, the point P on the sample 9 is projected onto the projection point R ′ on the Ronchigram by the electron beam 20 passing through the trajectory 221b. At this time, the movement of the projection point by defocus is given by the following equation [Equation 5].

Here, λ is given by the following equation.
[Formula 6]

However, in Expression 6, the defocus δε is left as small as the first order term.

In the equation, χ is the wavefront aberration function of Equation 1, and the parenthesized parenthesis represents the differentiation shown below.
[Formula 7]

Here, when calculating specifically by substituting Equation 1 for the aberration function χ of Equation 6, and assuming that the convergence angle (α, β) of the electron beam 20 is also sufficiently small, calculating up to its primary term , Equation 8 or Equation 9 is obtained.
[Formula 8]

[Formula 9]

By the way, if the displacement δ of the projection point on the Ronchigram is measured, the defocus C ₁ = C ₁ (α, β) with respect to the convergence angle (α, β) and A ₁ = A ₁ ( α, β) can be determined. However, as is clear from Equation 8 and Equation 9, undetermined coefficients to be determined, and C _1, the real and imaginary parts of A ₁ whereas it is three (Re A _1, Im A _1), there There are two independent equations. For this reason, all undetermined coefficients cannot be determined as they are.

Therefore, the same measurement is performed for the different defocus C ₁ ′ given by Equation 10.
[Formula 10]

Furthermore, if a displacement δ ′ given by Equation 11 described later is added, a sufficient number of equations can be obtained to determine all undetermined coefficients. Here, the focus change ε is a constant arbitrarily given as a prescribed amount from the outside.
[Formula 11]

As described above, if the focal point of the electron probe is slightly changed in the vicinity of the sample 9 during imaging of one Ronchigram under two defocus conditions, all unknown coefficients are determined from one Ronchigram. A sufficient number of equations can be obtained.

[Example 1]
FIG. 4 illustrates a partial configuration of the aberration correction STEM according to the first embodiment. In FIG. 4, the same reference numerals are given to corresponding parts to FIG. 1, and the configuration below the spherical aberration corrector 11 in the basic configuration shown in FIG. 1 is shown. Therefore, the configuration above the spherical aberration corrector 11 in the aberration correction STEM according to the present embodiment is the same as that in FIG. In FIG. 4, for the sake of simplification, the illustration of the bright field detector 14 and the dark field detector 13 which are not necessary for the description of the aberration correction using the Ronchigram is omitted.

The aberration correction control system 18 is provided with an aberration measurement system 181 that performs aberration measurement. The aberration measurement system 181 controls the lens current modulator 172 and the imaging detector 16 that modulate the excitation of the objective lens when performing aberration measurement. At the time of aberration measurement, the electron microscope control system 17 appropriately selects values such as the convergence angle of the convergent electron beam (electron probe) 21 with respect to the sample surface and the focal length of the objective lens 10, so that the optical conditions suitable for acquiring the Ronchigram are obtained. Set. Next, the electron microscope control system 17 projects the Ronchigram on the projection surface 15 after stopping and controlling the operation of the scan coil 24.

Here, the projection magnification (that is, the effective camera length) can be adjusted by the projection lens 12. The projected Ronchigram is acquired by the imaging detector 16 disposed on the back side (downward) of the projection surface 15. As described above, the imaging detector 16 acquires a Ronchigram under the control of the aberration measurement system 181. Further, in synchronism with the acquisition of the Ronchigram, the aberration measurement system 181 controls the lens current modulator 172 to apply a current that modulates excitation to the objective lens.

That is, in synchronism with the acquisition of one Ronchigram, the focal point of the convergent electron beam 21 is minutely displaced (vibrated) in the height direction in the vicinity of the sample 9. Due to the minute displacement of the focal point, as described with reference to FIG. 3, the positional change of the projection point R is captured in a form that is integrated into one Ronchigram (focus modulation Ronchigram). It is preferable that the minute displacement of the focus is periodically performed once or a plurality of (an integer of 1 or more) times within the acquisition time of the focus modulation Ronchigram.

[Example 2]
FIG. 5 shows a partial configuration of the aberration correction STEM according to the second embodiment. Also in the case of FIG. 5, like FIG. 4, only a part of the apparatus configuration is shown. In FIG. 5, the same reference numerals are given to the portions corresponding to FIG. 4. In the case of this embodiment, the minute displacement of the focal point is not performed by the excitation of the objective lens 10 but by the minute movement in the height direction of the sample 9 itself. Therefore, in the case of the present embodiment, the sample high displacement actuator 184 is attached to the sample holder 8 that holds the sample 9. As the sample high displacement actuator 184, for example, a piezoelectric element or the like is used. The sample high displacement actuator 184 in this embodiment slightly displaces the support portion of the sample holder 8 in the vertical direction.

Also in the case of the present embodiment, when the Ronchigram is acquired, the sample 9 is slightly displaced in the vertical (height) direction under the control of the aberration measurement system 181. Since the vertical displacement of the sample 9 is very small, for example, the actuator 184 may be a rotary type and the sample holder 8 may be slightly rotated around the axis to generate a local sample height displacement. Thus, the focal point can be changed even by changing the height of the sample itself. Also in the case of the present embodiment, similarly to the first embodiment, the focus is changed in one to several cycles in synchronization with the acquisition of the Ronchigram, and the focus modulation Ronchigram is acquired in an integrated manner.

[Example 3]
FIG. 6 illustrates a partial configuration of the aberration correction STEM according to the third embodiment. In the case of FIG. 6 as well, only a part of the apparatus configuration is shown as in the case of FIG. 4 and FIG. In FIG. 6, the same reference numerals are given to the portions corresponding to FIG. 4. In the case of the present embodiment, the focus adjustment lens 101 for measurement for causing a minute focus fluctuation is arranged above the objective lens 10, and this is controlled by the aberration measurement system 181 to execute the focus fluctuation. The aberration measurement system 181 executes the focus change of one to several cycles using the measurement focus adjustment lens 101 in synchronization with the acquisition of the Ronchigram, and realizes the acquisition of the integrated focus modulation Ronchigram.

In the case of the first embodiment (FIG. 4), the objective lens 10 itself to be corrected is used for the measurement although it is minute, so that lens drift, hysteresis, and the like affect the optical system of the STEM. The possibility is slight but worried. On the other hand, in the case of Example 2 (FIG. 5), it is possible to acquire a focus modulation Ronchigram without touching the STEM optical system such as the objective lens 10 and the spherical aberration corrector 11. Also in the case of the present embodiment (FIG. 6), the objective lens 10 itself does not need to change the excitation for measurement as in the second embodiment. In addition, the focus adjustment lens for measurement 101 used in this embodiment may be a very weak lens, and the influence of the aberration is almost negligible. Further, when an electrostatic lens is used as the measurement focus adjustment lens 101, for example, the influence can be completely removed from the optical system by turning off the measurement focus adjustment lens 101.

Since the Ronchigram is a long-distance projection image of the sample by the electron probe having a focal point in the vicinity of the sample as described above, this can be expressed by a probe shape and convolution on the sample surface as follows.
[Formula 12]

Here, A (α, β) is a limitation of the electron beam that contributes to the formation of the electron probe, that is, a so-called aperture function. 1 + Ψ (x, y) is a function representing a phase change due to the sample 9. Sample 9 used weak phase object approximation.

In Expression 12 indicating the formation of the Ronchigram, introducing defocus ε is expressed as Expression 13 when this is included in the aberration function and integrated with respect to defocus. The integration range of Equation 13 is {−∞, + ∞}.
[Formula 13]

Here, σ (ε) is a distribution function related to the focus distribution (focal blur), and is given by Equation 14 if a Gaussian distribution with a width δε is assumed, for example.
[Formula 14]

Further, δ (ε) is given by Equation 15 when considering the case of using a sin wave with alternating current.
[Formula 15]

[Evaluation of aberration by focus modulation Ronchigram obtained in each example]
FIG. 7A shows two types of Ronchigrams obtained when the amorphous thin film 90 is used as the sample 9. FIG. 7B is a normal Ronchigram 151 and FIG. 7C is a focus modulation Ronchigram 152. Both the normal Ronchigram 151 and the focus modulation Ronchigram 152 are formed by connecting an electron probe from a Gaussian focus to ε ₁ = + 50 nm. Further, in the case of the focus modulation ronchigram, a focus variation of δε = ± 10 nm is added around this ε ₁ = + 50 nm. In addition, various aberration amounts are listed outside the table.

In the case of the method described in Patent Document 1 described above, the grain shape of a local amorphous projection image is evaluated from a normal Ronchigram 151 using a fitting operation such as an elliptic function, and axial aberrations are obtained using these local aberration information. Is calculated. In

Patent Documents

2 and 3, two Ronchigrams 151 are obtained before and after the positional relationship between the focal point and the sample surface is slightly changed in the height direction, and axial aberration is obtained using local displacement between corresponding regions as local aberration information. Is calculated.

On the other hand, in the case of each of the above-described embodiments, the focus modulation Ronchigram 152 is used, and the locus of the amorphous particles projected in one focus modulation Ronchigram 152 (the displacement amount due to the minute fluctuation of the focus) is used. And local aberration information. These follow λ in Equations 6-8. Of course, the aberration correction STEM (FIGS. 4 to 6) and the driving method having the above-described configuration are used to obtain the focus modulation Ronchigram 152.

8A to 8D, a specific method for extracting local aberration information will be described. First, an area suitable for measurement is extracted from the focus modulation Ronchigram 152 (FIG. 8A). In the figure, the area to be extracted is shown surrounded by a white frame. Next, the area is divided into appropriate sizes (FIG. 8B). Here, an example of dividing into 5 × 5 grids is shown. However, the number of divisions of the grid is not limited to 5 × 5, and it is not necessary to divide into a grid in the first place. Moreover, a part of each division area may mutually overlap.

Next, autocorrelation is taken for each region after the division in order to clarify the elongation strain amount of the particles (FIG. 8C). Note that the method for clarifying the elongation and distortion amount of each area after division is not limited to the method of obtaining the autocorrelation of each area. For example, the focus modulation ronchigram of the sample-specific part imaged under the same acquisition conditions may be similarly divided into regions, and cross-correlation may be taken between the corresponding divided regions. Further, the cross-correlation may be calculated between the corresponding partial (divided) regions between the focus modulation Ronchigram and the normal Ronchigram 151 (FIG. 7B) acquired at the same sample location.

Finally, the amount of elongation strain in each divided region is quantified, and a line element indicating the length and direction is obtained in each divided region (FIG. 8D). The x-y component of this line element corresponds directly to Expression 7, or the length and direction correspond to Expression 8. Therefore, Equation 7 or Equation 8 is established from these measured values. Further, as described in the explanation of FIG. 3, the shortage of the simultaneous equations is obtained in the same way by obtaining the focal modulation Ronchigram at different focal positions ε ₂ (≠ ε ₁ ), compensate. As a result, the defocus C ₂ and astigmatism A ₂ for each divided region can be obtained. Further, by substituting these values into the following equation 16, a simultaneous equation up to the required aberration order is established, and the axial aberration is calculated.
[Formula 16]

Here, (α _i , β _i ) is the convergence angle of the electron beam with respect to the center of the i-th divided region, for example, that shown in FIG.

The elongation strain measurement process corresponding to FIGS. 8C and 8D can be realized by a simple method as shown in FIG. 9, for example. That is, the maximum value and its position for each column (row) in one direction are searched and plotted for one divided area. For example, in the example of FIG. 9, the position (pixel) that gives the maximum value in each column is found from the correlation image 1520, and this is indicated by a broken line 1522 connecting A to B. Further, the correlation strength measured along this broken line Is plotted in the lower graph 1521. By referring to the upper and lower graphs and selecting pixels having correlation strengths equal to or higher than a threshold value determined appropriately and linearly approximating these positions, a line element 1524 indicated by a bold line can be determined. If this is carried out in each divided region, a line element 1524 representative of elongation strain can be obtained similarly, that is, defocus C ₁ and astigmatism A ₁ can be obtained in each region.

Using the above method, focus modulation Ronchigrams are acquired and analyzed at two focal positions. Thereby, the axial aberration coefficient of the part related to the formation of the STEM electron probe including the objective lens 10 and the spherical aberration corrector 11 can be obtained.

[Features of the embodiment]
The first feature of the measurement method according to each embodiment is that, as in Patent Document 2 and Patent Document 3, two Ronchigrams are not required before and after the focus change, and the focus is 1 during acquisition of one Ronchigram. The focus modulation Ronchigram can be acquired by one step of changing a plurality of periods.

The second feature of the measurement method according to each example is that, as in Patent Document 2, it is not necessary to track a specific point in the sample with a Ronchigram before and after the focus change. In the measurement method proposed in this specification, since the pattern of each region image obtained by dividing the focus modulation Ronchigram appears according to the movement of each sample particle, the amount of elongation distortion of the region pattern using autocorrelation as in the above method Can be measured.

The third feature of the measurement method according to each embodiment is that the detection sensitivity of the local aberration information can be adjusted with a focus fluctuation amount δε that can be arbitrarily controlled. For example, in the case of FIG. 8D, the inner 3 × 3 region has a shorter line element length that gives local aberration information than the outer peripheral region, and seems to easily cause an error. Since the line element length is considered to be proportional to δε from Equation 9, in such a case, a focus variation Ronchigram in which δε is increased in the same field of view should be obtained. By combining a plurality of focus modulation ronchigrams having different δε and combining data of δε that can obtain an appropriate line element length in each divided region, improvement in measurement accuracy can be expected.

[Effect of Example]
As described above, by adopting the aberration measurement method using the focus modulation Ronchigram, it is possible to provide a high-precision aberration measurement method capable of measuring on-axis aberration coefficients up to higher orders, particularly in the aberration correction STEM.

Further, in the aberration measurement method of the present embodiment, the Ronchigram is used for measurement, so that each stage is different from the probe tableau method described in Non-Patent Document 1 (aberration measurement method that requires repeated measurement conditions). It is possible to measure aberrations with a simplified measurement procedure. Accordingly, it is possible to expect a significant reduction in the time required for aberration measurement.

Also, in the conventional aberration measurement method using Ronchigram (Patent Document 2 and Patent Document 3), a plurality of specific observation points are determined on the sample before and after the change of measurement conditions such as defocus and beam shift, and the displacement of the same location is determined. Must be tracked and local aberration information extracted. However, in the case of the aberration measurement method of this embodiment, it is not necessary to set a specific observation point in the sample, and the necessary local aberration information is extracted from the form of the focus modulation Ronchigram itself to calculate the axial aberration coefficient. Can do. The focus modulation Ronchigram used at this time is an image obtained by performing focus modulation of one to a plurality of periods within the photographing time of one Ronchigram as described above, and obtained before and after the change of the measurement condition as described above. There is no need to compare multiple Ronchigrams.

The aberration measurement method of the present example has the same advantages as the measurement method described in Patent Document 1. In the aberration measurement method of Patent Document 1, a Ronchigram is also used for the measurement, and this is divided into small areas, the autocorrelation of each area image is measured, and local aberration information is extracted from each. However, the aberration measurement method of Patent Document 1 requires work such as fitting the autocorrelation strength obtained in each region with an elliptic function.

However, fitting errors and autocorrelation in the local area of the Ronchigram tend to show sub-peaks in addition to the main peak. Care must be taken to extract correct local aberration information from the autocorrelation using the method described above. . On the other hand, according to the aberration measurement method of the present embodiment, the focus modulation Ronchigram is divided into local regions as in Patent Document 1, and this autocorrelation or cross-correlation is taken to extract local aberration information from each. All that is required is the length and orientation of the main peak that appears in the correlation pattern of each region. This can be accurately obtained by the simple method described in FIG.

In this regard, the length of the main peak of this correlation pattern is proportional to the amount of focus modulation at the time of acquiring the focus modulation Ronchigram, as shown in Equation 8 or Equation 9. Utilizing this property, when the main peak length is not long enough for measurement (when it is smaller than the threshold value) or too long (when larger than the threshold value) depending on the aberration state The aberration measurement system 181 can guarantee the measurement accuracy by adjusting the focus modulation amount through the electron microscope control system 17 and optimizing the main peak length to be measured.

Although the number of times of measurement will increase, for example, if a plurality of Ronchigrams in which the amount of focus modulation is adjusted so that a correlation pattern appropriate for the measurement of local aberration information is obtained for each region into which the focus modulation Ronchigram is divided are used. Further, the measurement accuracy of the local aberration information can be further improved.

In addition, in the aberration measurement method of Patent Document 1, the sample to be measured is limited to a pure amorphous thin film. On the other hand, in the aberration measurement method of the present embodiment, it is only necessary to know the displacement trajectory of the particle according to the focus modulation. Therefore, if it does not affect this, the measurement sample includes substances other than amorphous, such as crystal fine particles. It may be.

[Other embodiments]
The present invention is not limited to the configuration of the embodiment described above, and includes various modifications. For example, in the above-described embodiments, the description has been made on the premise of STEM, but other charged particle optical devices (for example, scanning electron microscope (SEM), focused ions, etc.) that scan with the charged particle probe focused on the surface of the observation sample Even in the case of a beam (FIB) processing machine, the present invention can be applied when a Ronchigram can be obtained by devising a measurement sample.

The present invention can also be applied to an STEM that does not include the spherical aberration corrector 11. In that case, the measured axial aberration cannot be used for aberration correction, but the aberration of the objective lens 11 used in the STEM can be accurately evaluated and used for later use.
In the above description, the case where the above-described aberration measuring method is applied regardless of the type of the sample has been described. However, an aberration measurement method applied to each sample may be registered as a recipe in advance. A function for automatically generating a recipe may be mounted on the aberration correction STEM.

In addition, the above-described embodiments are described in detail for some embodiments in order to explain the present invention in an easy-to-understand manner, and it is not always necessary to provide all the configurations described. Further, a part of a certain embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of a certain embodiment. It is also possible to add other configurations to the configuration of each embodiment, replace a partial configuration of each embodiment with another configuration, or delete a partial configuration of each embodiment.

In addition, each of the above-described configurations, functions, processing units, processing means, and the like may be partly or entirely realized as, for example, an integrated circuit or other hardware. Each of the above-described configurations, functions, and the like may be realized by a processor interpreting and executing a program that realizes each function. That is, each configuration may be realized by software. In this case, information such as programs, tables, and files for realizing each function can be stored in a storage device such as a memory, a hard disk, an SSD (Solid State Drive), or a storage medium such as an IC card, an SD card, or a DVD. .

Also, the control lines and information lines indicate what is considered necessary for explanation, and do not represent all control lines and information lines necessary for the product. In practice, it can be considered that almost all components are connected to each other.

1: mirror body, 2: accelerator tube, 3: electron source, 4: converging lens, 5: converging lens diaphragm, 6: deflector, 7: converging lens, 8: sample holder, 9: sample, 10: objective lens, 11: spherical aberration corrector, 12: projection lens, 13: dark field detector, 14: bright field detector, 15: projection surface (fluorescent screen), 16: imaging detector (CCD camera, etc.), 17: electron microscope control System: 18: Aberration correction control system, 181: Aberration measurement system, 182: Sample high displacement power supply, 184: Sample high displacement actuator, 19: Control computer (control PC), 20: Electron beam, 21: Convergent electron beam 21a, 210a: electron beam trajectory (approximate trajectory) converging in an ideal state on the sample surface, 21b, 210b: electron beam trajectory converging with aberration on the sample surface, 211a: convergence in the ideal state after focal displacement Electric Line trajectory (approximate trajectory), 211b: trajectory of electron beam that converges with aberration after focal displacement, 22a, 220a: trajectory of electron beam that passes through the sample in an ideal state (approximate trajectory), 22b, 220b: specimen trajectory Trajectory of electron beam that passes through aberration, 221a: Trajectory of electron beam that passes through in ideal state after focal displacement (approximate trajectory), 221b: Trajectory of electron beam that passes through aberration after focal displacement, 20Ta: Irradiated electron Line (perpendicular incidence), 20Tb: irradiation electron beam (gradient incidence), 24: scan coil, 30a: ideal wavefront of electron beam, 30b: electron wavefront including aberration, 90: sample (amorphous thin film), 101: focus for measurement Adjustment lens, 150: projection plane, 151: Ronchigram image, 152: focus modulation Ronchigram image, 171: objective lens power supply, 172: lens current modulation power supply, 252: An image obtained by dividing a measurement region from a point modulation Ronchigram, an image showing the autocorrelation strength of each region of 253: 252, an image showing the amount of elongation distortion of the autocorrelation strength of each region of 254: 253 quantified by a linear element, 1520 : Example of correlation intensity image of one region, 1522: Polygonal line connecting points indicating the highest intensity in each column of the correlation intensity image, 1523: Correlation intensity profile along broken line AB, 1524: Quantification of elongation distortion amount of correlation intensity Line element.

Claims

An electron source for generating an electron beam;
Accelerating means for accelerating the electron beam to a predetermined energy;
A lens or lens group for converging the accelerated electron beam;
An objective lens that forms an image of an electron probe on the surface of an observation sample disposed on the downstream side of the lens or lens group;
A scanning system that two-dimensionally scans the electron probe within a specified range in the surface direction of the observation sample;
An imaging unit that detects electrons transmitted through the observation sample and acquires a Ronchigram;
A control unit that operates in synchronization with the imaging unit, and obtains the relationship between the focus of the electron probe imaged on or near the surface of the observation sample and the height of the observation sample during acquisition of one Ronchigram A scanning transmission electron microscope having a control unit that minutely fluctuates.
The scanning transmission electron microscope according to claim 1,
The scanning transmission electron microscope characterized in that the control unit minutely varies the height relationship by minutely varying the excitation current of the objective lens.
The scanning transmission electron microscope according to claim 1,
The scanning transmission electron microscope characterized in that the control unit minutely varies the height relationship by minutely varying the mounting height of the holder that holds the observation sample.
The scanning transmission electron microscope according to claim 1,
The scanning transmission electron microscope according to claim 1, wherein the control unit minutely varies the height relationship by minutely varying excitation of a focus adjusting lens different from the objective lens.
The scanning transmission electron microscope according to claim 1,
Based on the aberration information measured from the one Ronchigram acquired in synchronism with the focus variation of one or more periods, the axial aberration of the electron probe imaging system including the objective lens is calculated to a higher order. A scanning transmission electron microscope further comprising a measurement unit.
The scanning transmission electron microscope according to claim 5,
The measuring unit measures a local elongation strain amount from each of a plurality of local regions in the one Ronchigram, and calculates the on-axis aberration based on the plurality of local elongation strain amounts. electronic microscope.
The scanning transmission electron microscope according to claim 5,
When the amount of local elongation strain measured from each of a plurality of local regions in the single Ronchigram does not have an appropriate length, the control unit may increase the height so that the amount of local elongation strain satisfies the appropriate length. The scanning transmission electron microscope is characterized in that the measurement unit measures the amount of local elongation strain based on a plurality of Ronchigrams acquired through adjustment of the height relationship.
The scanning transmission electron microscope according to claim 5,
The measurement unit measures a local elongation strain amount of the first region from the first Ronchigram acquired by the first variation of the height relationship, and is acquired by the second variation of the height relationship. A scanning transmission electron microscope characterized in that a local elongation strain amount in the second region is measured from the second Ronchigram.
An electron source for generating an electron beam; acceleration means for accelerating the electron beam to a predetermined energy; a lens or a lens group for converging the accelerated electron beam; and a downstream side of the lens or the lens group. In an aberration measurement method in a scanning transmission electron microscope having an objective lens that forms an electron probe on the surface of an observation sample and a scanning system that two-dimensionally scans the electron probe within a specified range in the surface direction of the observation sample.
A process in which the control unit minutely varies the relationship between the focus of the electron probe imaged on or near the surface of the observation sample and the height of the observation sample;
A process in which an imaging unit that operates in synchronization with the control unit acquires a single Ronchigram that includes change information of a projected image due to minute variations in the height relationship in the image;
And a process of measuring aberration information from the one Ronchigram.
The aberration measurement method according to claim 9,
The aberration measurement method, wherein the control unit minutely varies the height relationship by minutely varying an exciting current of the objective lens.
The aberration measurement method according to claim 9,
The aberration measuring method, wherein the control unit minutely varies the height relationship by minutely varying a mounting height of a holder for holding the observation sample.
The aberration measurement method according to claim 9,
The aberration measurement method, wherein the control unit minutely varies the height relationship by minutely varying excitation of a focus adjusting lens different from the objective lens.
The aberration measurement method according to claim 9,
The measurement unit increases the on-axis aberration of the electron probe imaging system including the objective lens based on the aberration information measured from the one Ronchigram acquired in synchronization with the focus variation of one or more periods. An aberration measurement method characterized by calculating to the following.
The aberration measurement method according to claim 13,
The measurement unit measures local elongation strain amounts from a plurality of local regions in the one Ronchigram, and calculates the axial aberration based on the plurality of local elongation strain amounts. Method.
The aberration measurement method according to claim 13,
When the amount of local elongation strain measured from each of a plurality of local regions in the single Ronchigram does not have an appropriate length, the control unit may increase the height so that the amount of local elongation strain satisfies the appropriate length. And the measurement unit measures the amount of local elongation distortion based on a plurality of Ronchigrams acquired through adjustment of the height relationship.