JP5347496B2 - Method for measuring chromatic aberration coefficient in electromagnetic lens and scanning transmission electron microscope - Google Patents

Description

  The present invention relates to a method for measuring a chromatic aberration coefficient in an electromagnetic lens and a scanning transmission electron microscope, for example, a configuration for simply and accurately measuring a chromatic aberration coefficient of an electromagnetic lens of a scanning transmission electron microscope equipped with a spherical aberration correction device. It is about.

  In high-resolution image observation with a scanning transmission electron microscope (STEM), since the spherical aberration coefficient of the electromagnetic lens is very large, sufficient resolution cannot be obtained with respect to the wavelength of the electron beam. However, in recent years, by mounting a spherical aberration correction device on an electron microscope, the influence of spherical aberration has become almost negligible, and it has become possible to sufficiently bring out the performance of the objective lens.

  However, on the other hand, the spherical aberration corrector has the disadvantage of increasing the chromatic aberration coefficient. Recently, a chromatic aberration coefficient correction apparatus has been developed for solving the problem, but the chromatic aberration correction apparatus for electromagnetic lenses has a feature that it is greatly different from an optical lens.

  Regarding such chromatic aberration correction devices for electromagnetic lenses, TEAM, a national project in the United States, conducted joint research with FEI, while in Japan, Dr. Kazutomo Suenaga of National Institute of Advanced Industrial Science and Technology developed with JEOL. Has started. However, as of September 24, 2008, no announcement has been made regarding the results of chromatic aberration correction. Furthermore, there is no established method for measuring the chromatic aberration coefficient with high accuracy.

As a conventional method for measuring the chromatic aberration coefficient, a method of calculating the chromatic aberration coefficient using a simulation based on the lens shape is known. In addition, a method has been proposed in which the relationship between the change in the acceleration energy of the electron beam and the change in chromatic aberration is obtained from several diffraction images, and the chromatic aberration coefficient is calculated from the relational expression (for example, see Patent Documents 1 to 3) ).
JP 2006-85919 A JP 2006-114303 A JP 2007-180013 A

  However, the above-described simulation method based on the lens shape can be applied to a system in which the lens shape does not change, such as an optical microscope, but the lens shape itself changes depending on measurement conditions, such as an electromagnetic lens. It is not suitable for such cases. Furthermore, since the lens configuration of the spherical aberration correction device changes depending on the correction conditions, there is a problem that it is very difficult to measure the chromatic aberration coefficient under actual measurement conditions.

  On the other hand, the method of obtaining the chromatic aberration coefficient from the above-described diffraction image has a problem that a large amount of error is included because the amount of defocus is changed while the acceleration voltage is changed.

  Accordingly, an object of the present invention is to easily and accurately measure a chromatic aberration coefficient by simply obtaining a single high-resolution electron microscope image without changing the acceleration voltage of electrons.

  From one aspect of the present invention, a high-resolution scanning transmission electron having a resolution of 0.3 nm or less is obtained by making a convergent electron beam that has passed through an electromagnetic lens corrected for spherical aberration incident on a single crystal sample having a known crystal structure and lattice constant. A step of obtaining a microscope image by actually measuring a microscope image obtained by continuously changing the amount of defocus of the electromagnetic lens when obtaining a microscope image, and a regression curve for obtaining a regression curve from an intensity profile obtained by averaging the microscope image in the scanning direction. Chromatic aberration coefficient for calculating the chromatic aberration coefficient of the electromagnetic lens by comparing the chromatic aberration coefficient obtained from the calculation process of the theoretical microscope image obtained by the theoretical calculation under the same conditions as the microscope image and the regression curve and the parameters of the regression curve A method for measuring a chromatic aberration coefficient in an electromagnetic lens having a calculation step is provided.

  According to another aspect of the present invention, an electron beam generating unit that generates an electron beam, a converging lens that converges the electron beam generated by the electron beam generating unit, and a spherical aberration that corrects the spherical aberration of the converging lens A correction unit; a scanning coil unit that scans a convergent electron beam applied to the sample; a detector that detects a diffraction image intensity on the diffraction image surface of the sample; and a display unit that images the detection intensity of the detector A control unit that changes the current value of the convergent lens as the amount of defocus of the sample, and a calculation unit that calculates an average intensity profile of the acquired diffraction image, and a calculation result and theory by the calculation unit A scanning transmission electron microscope provided with a chromatic aberration coefficient measurement mechanism of an electromagnetic lens having at least chromatic aberration coefficient calculation means for calculating a chromatic aberration coefficient in comparison with a calculation result is provided.

  According to the disclosed chromatic aberration coefficient measuring method and scanning transmission electron microscope in an electromagnetic lens, it is possible to easily and accurately measure a chromatic aberration coefficient simply by acquiring a single microscope image without changing the acceleration voltage of electrons. It becomes possible.

Here, an embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a conceptual apparatus configuration diagram of a scanning transmission electron microscope provided with a chromatic aberration measuring mechanism according to an embodiment of the present invention. The scanning transmission electron microscope includes an electron gun 11, an irradiation lens 13, a spherical aberration correction device 14, an objective lens 15, a STEM detector 17, a display unit 18, and a control unit 19. The sample 16 may be any single crystal, but is preferably a SrTiO ₃ single crystal or a Si single crystal from the viewpoint of availability.

  In this case, the spherical aberration correction device 14 for the irradiation lens is located between the irradiation lens 13 and the objective lens 15, and the spherical aberration of the irradiation lens 13 is such that the electron beam of the same energy converges at one point above the sample 16. It is corrected. The electron beam 12 scans the sample 16 by a scanning coil (not shown) located above the objective lens 15. The objective lens 15 is used to adjust the focus of the formed convergent electron beam.

  The intensity on the diffraction image plane is detected by the STEM detector 17 located below the sample 16, and the detection signal is displayed on the display unit 18 in real time in synchronization with the electron beam scanning. Further, according to the signal from the control unit 19, the objective lens 15 changes the current value so as to change the focal position. The functions of the control unit 19 and the display unit 18 constitute a chromatic aberration coefficient measurement mechanism.

  FIG. 2 is a schematic explanatory view of changes in the electron beam trajectory when the electromagnetic lens includes chromatic aberration. If the energy of the incident electron beam 22 is completely aligned (ΔE = 0), the electromagnetic lens 21 is changed. The passing electron beam follows the same trajectory. However, in the current electron microscope, the energy of the electron beam 22 spreads due to various factors. The spread of energy can be rephrased as having a range of wavelengths.

  Since the refraction angle when passing through the electromagnetic lens 21 varies depending on the wavelength, the electron passing through the electromagnetic lens 21 is blurred at the arrival point by the amount corresponding to the wavelength width. This blur affects the high resolution image as it is.

Here, the divergence angle of the electron beam 22 incident on the electromagnetic lens 21 is α, the defocus amount is Δf, the blur on the image plane 23 caused by the defocus is r, the energy of the electron beam 21 is E, and the variation is ΔE, If the chromatic aberration coefficient is C _c ,
r = C _c α (ΔE / E)
There is a relationship. At this time, as apparent geometrically, the defocus amount Δf and the blur r are in a proportional relationship, so the defocus amount Δf and the chromatic aberration coefficient C _c are also in a proportional relationship.

FIG. 3 is a theoretical STEM image in the case of using a SrTiO ₃ single crystal having a (001) plane before and after chromatic aberration correction as a sample, and FIG. 3A is a STEM image before chromatic aberration correction. b) is a STEM image after chromatic aberration correction. From the comparison between FIG. 3A and FIG. 3B, it is clear that the STEM image is blurred depending on the presence or absence of chromatic aberration. In the theoretical calculation of the STEM image, the calculation is performed on the assumption that the resolution of the scanning transmission microscope is 0.3 nm or less, that is, the resolution is high enough to identify atoms.

  Therefore, it is possible to accurately quantify the influence of blur by measuring the chromatic aberration coefficient, and it is possible to improve the accuracy of quantitative measurement by incorporating it into the theoretical calculation.

FIG. 4 is an explanatory view of the defocus amount dependency of contrast with respect to each chromatic aberration coefficient by theoretical calculation. Here, the amount of defocus Δf is 0 nm, −10 nm, and −20 nm, and the chromatic aberration coefficient C _c is calculated using values of 0 mm, 1.0 mm, and 2.0 mm. Incidentally, chromatic aberration coefficient C _c before the spherical aberration correction of the current scanning transmission electron microscope is about 2.0 mm.

As can be seen from FIG. 4, when the chromatic aberration coefficient is 0 mm, the defocus amount is about −10 nm and the image contrast is reduced, whereas when the chromatic aberration coefficient is 2.0 mm, the defocus amount is − Even if it changes by 20 nm, the contrast of the image remains. From this result, the chromatic aberration coefficient C _c and the defocus amount Δf are correlated, it is possible to measure the chromatic aberration coefficient C _c if it is possible to obtain the relationship between contrast and chromatic aberration coefficient C _c of the diffraction image.

  FIG. 5 is an explanatory diagram of an STEM image, FIG. 5A is an explanatory diagram of an STEM image acquisition process, and FIG. 5B is an STEM image of a Si single crystal having a (011) plane as a main surface. It is. As shown in FIG. 5A, in order to acquire a STEM image, the converging electron beam 31 is scanned two-dimensionally on the upper part of the sample 32, and the scattering intensity is detected by a detector. In FIG. 5A, the scanning direction indicated by the arrow is the main scanning direction, and the direction orthogonal to the arrow is the sub-scanning direction.

  FIG. 5B is a STEM image of an Si single crystal having the (011) plane as a main surface, which was actually photographed, and is shown by (A), (B), and (C) shown in FIG. Images are acquired as the intensities of the positions of the converged electron beams 31 corresponding to the positions (A) to (C) of FIG. That is, not only the two-dimensional images are photographed under the same conditions, but the experimental conditions can be changed corresponding to the position of the convergent electron beam during scanning.

Therefore, in the embodiment of the present invention, continuous in order to examine the correlation of the contrast of the chromatic aberration coefficient C _c and STEM image, when acquiring a high-resolution STEM images discernible single atom, a defocus amount Δf The relationship between the contrast of the image and the amount of defocus is examined by changing at regular intervals.

  FIG. 6 is a continuous defocus change STEM image, which is based on the theoretical calculation result of the chromatic aberration coefficient dependence of the contrast of the STEM image. The STEM image of the Si single crystal which makes the (011) plane the main surface when the amount of defocus Δf is continuously changed from −100 nm to 100 nm is shown. Here, in the sub-scanning direction, the defocus amount Δf is changed by 0.78 nm with a pitch of 0.048 nm.

  In addition, as a change range of the defocus amount Δf, a range showing a change that can be obtained in a regression curve described later is required, and even when the chromatic aberration coefficient is 4.0 mm, which is twice the current about 2.0 mm. In order to obtain the regression curve, the above-mentioned range of −100 nm to 100 nm is required.

  This is because, in order to accurately evaluate an error of a value obtained by linear approximation, sample points widely distributed before and after an expected value are necessary. Note that the chromatic aberration coefficient decreases when hardware improvements such as further reducing the energy variation ΔE of the electron beam from the electron gun are made, so the focus may be shifted in a range from -100 nm to less than 100 nm.

FIG. 6A to FIG. 6C show the results calculated with the chromatic aberration coefficient C _c being 0.0 mm, 2.0 mm, and 4.0 mm, respectively. The smaller the chromatic aberration coefficient C _c contrast it can be seen that sharply weak against defocus amount Delta] f. Intensity profiles obtained by averaging these image intensities in the scanning direction are shown by overlapping with white lines.

  In the present invention, a regression curve parameter obtained from the maximum value of the intensity profile is used as a parameter representing contrast. Here, a parameter obtained when the regression curve is a normal distribution function is used, and typically the full width at half maximum or half width at half maximum of the normal distribution function is used.

FIG. 7 is an explanatory diagram of the correlation between the full width at half maximum of the normal distribution function and the chromatic aberration coefficient. Here, the sample has a thickness t of, for example, two (011) planes of 40 nm and 80 nm as principal surfaces. a Si single crystal, shows a full width at half maximum obtained spherical aberration coefficient _{C c} as 0.005mm and 0.0005 mm. Symbols ○ is shown in FIG. 7 shows the case of 0.005mm to _{C c} at t = 40 nm, the symbol □ indicates the case of 0.0005mm the _{C c} at t = 40 nm, the symbol × is t = 80 nm in _{C c} Is 0.0005 mm.

As apparent from FIG. 7, most results unchanged in any condition, the relationship of the chromatic aberration coefficient C _c and FWHM t is a linear relationship. From this, it may be stored a relationship of the chromatic aberration coefficient C _c and FWHM t in a database seeking in a standard each sample in the memory of the control unit 19 described above, the chromatic aberration by substituting experimental values in this equation The coefficient _Cc can be easily calculated.

Further, from these results, the chromatic aberration coefficient measurement method according to the embodiment of the present invention is insensitive to the thickness of the sample, and the spherical aberration coefficient C _c is also corrected within the range corrected by the spherical aberration correction device. It can also be seen that the aberration coefficient has little effect on the chromatic aberration coefficient measurement. Specifically, measurement is possible if the thickness of the sample is 100 nm or less and the spherical aberration coefficient is 0.005 mm or less.

Further, the accuracy of measurement in the embodiment of the present invention is defined by the error relation for obtaining the relationship between the contrast and the chromatic aberration coefficient C _c. In the measurement conditions described above, it can be measured chromatic aberration coefficient _{C c} is within the error range of ± 0.25 mm.

In the measurement by simulation based on conventional lenses engineering, it has not been possible to obtain the chromatic aberration coefficient C _c of the spherical aberration corrected electromagnetic lenses precisely. However, as described above, since the chromatic aberration coefficient measurement method according to the embodiment of the present invention can be measured with a crystalline sample that is actually tested, the chromatic aberration coefficient C _c under conditions closer to the actual measurement conditions. Can be measured accurately.

In addition, in the method of measuring using the energy of different electron beams that requires several STEM images, an error due to a spontaneous change in the defocus amount Δf is included during the measurement. However, the chromatic aberration coefficient measurement method according to the embodiment of the present invention, since it is possible to measure the chromatic aberration coefficient C _c by simply obtaining a single STEM image, there is a strong resistance to the natural change in defocus amount.

These advantages, in the conventional method can be easily measured chromatic aberration coefficient C _c is possible to accurately measure difficult. Also, quantification of device functions closely related to the resolution of the high resolution image also can be performed in After accurately measure the chromatic aberration coefficient C _c.

  Based on the above, a method for measuring a chromatic aberration coefficient of an electromagnetic lens according to Example 1 of the present invention will be described next with reference to FIG. FIG. 8 is a flowchart of the method for measuring the chromatic aberration coefficient of the electromagnetic lens of Example 1 of the present invention.

This flowchart can be roughly divided into two processes. One is processing by theoretical calculation on the right side of the flowchart. As shown in FIG. 6 above, in some of the chromatic aberration coefficient C _c, the acceleration voltage V (or, E = eV), high resolution combined incidence angle of convergence, the capture angle detectors STEM image in experimental conditions A continuous defocus change image of the STEM image is calculated. A regression curve is obtained from the calculation result using a normal distribution function, and a calibration curve for the chromatic aberration coefficient and the full width at half maximum of the regression curve is obtained using the full width at half maximum as a parameter for quantifying the contrast of the STEM image.

  If the experimental conditions in the scanning transmission electron microscope are limited, if these calibration results are left in the control unit as a database, the theoretical calculation part is not necessarily required for each measurement.

The other process is a process part by actual measurement on the left side of the flowchart. The conditions of the incident electron beam are the acceleration voltage V, the incident convergence angle of the incident electron beam, and the aberration condition of the electromagnetic lens. Further, the spherical aberration correction device is adjusted so that the spherical aberration coefficient is 0.005 mm or less.
At this time, when the calibration of the defocus amount Δf and the current value of the objective lens is performed, the spherical aberration correction device is turned off once, and the calibration of the current value and the defocus amount Δf is performed using a Ronchigram (for example, (See JP 2007-109509 A).

The calibration result is reflected in the control unit 19. In the control unit 19, the current value of the objective lens is changed during acquisition of one STEM image by an amount corresponding to the defocus amount Δf from −100 nm to 100 nm continuously and at regular intervals, and a continuous defocus change image is acquired. . Thereafter, as in the theoretical calculation portion of chromatic aberration by comparing to a calibration result sought full width at half maximum of the normal distribution function coefficient C _c is calculated.

A specific procedure of the method for measuring the chromatic aberration coefficient of the electromagnetic lens according to the first embodiment of the present invention will be described below based on the flowchart shown in FIG. 8, and the reference numerals in FIG. First, in step 1 (S1),
S1: A single crystal sample 16 is inserted into the STEM, and the condition is input to the control unit 19. Then
S2: The optical axis adjustment of STEM is performed and the condition is input to the control unit 19. Then
S3: A Ronchigram is acquired, and the control unit 19 uses the Ronchigram to calibrate the defocus amount Δf and the electromagnetic lens current amount. Then
S4: Activate the spherical aberration correction device. Then
S5: The spherical aberration correction device is adjusted and the condition is input to the control unit 19. Then
S <b> 6: The control unit 19 changes the current value of the objective lens 15 during acquisition of one STEM image, acquires the continuous defocus change image shown in FIG. 6, and inputs it to the control unit 19. Then
S7: As shown in FIG. 6, the control unit 19 performs an averaging process in the scanning direction to create an intensity profile. Then
S8: As shown in FIG. 6, the control unit 19 obtains a regression curve based on a normal distribution function based on the intensity profile, and obtains the full width at half maximum as a parameter.
The above is the process accompanying the actual measurement.

Next, in the control unit 19,
S9: The database of the calibration curve of chromatic aberration coefficient _Cc and contrast shown in FIG. As a result of the search, if there is no data that matches the sample condition, the process proceeds to the next step S10, if there is matching data and sample conditions calculates the chromatic aberration coefficient C _c jumps to step S16. If there is no data that matches the sample conditions, in the next step S10,
S10: An arbitrary chromatic aberration coefficient _Cc is input to draw a calibration curve of chromatic aberration coefficient _Cc and contrast. Then
S11: Based on the input chromatic aberration coefficient C _c , sample conditions, and experimental conditions, the high-resolution continuous defocus change image shown in FIG. 6 is calculated. Then
S12: An intensity profile is created by performing averaging processing in the scanning direction on the continuous defocus change image. Then
S13: As shown in FIG. 6, a regression curve by a normal distribution function is obtained based on the intensity profile, and the full width at half maximum is obtained as a parameter. Then
S14: It is determined whether or not a calibration curve of chromatic aberration coefficient _Cc and contrast can be drawn.

As a result of the determination, if the calibration curve can be drawn, the process proceeds to the next step S15. If the calibration curve cannot be drawn, the process returns to S11 and another chromatic aberration coefficient _Cc is input to generate the calibration curve. Steps S11 to S14 are repeated until it can be drawn. In order to draw a calibration curve, it is necessary calculation result by at least two of the chromatic aberration coefficient C _c, in order to reduce the error-variance of the calibration curve, for example, by 6 or more of the chromatic aberration coefficient C _c Calculate and draw a calibration curve.

As a result of determination, if a calibration curve can be drawn,
S15: The calibration result is registered in the database stored in the analysis data memory area of the control unit 19. Then
S16: The calibration curve data is compared with the full width at half maximum obtained by actual measurement in step S8, the chromatic aberration coefficient _Cc is calculated, and the series of flow ends.

  Note that the chromatic aberration coefficient measurement mechanism according to the first embodiment of the present invention has a function of performing the processes of steps S6 to S16. Actually, it is performed by the control unit 19, and the display unit 18 is also used as necessary.

It is a conceptual apparatus block diagram of the scanning transmission electron microscope provided with the chromatic aberration measuring mechanism of embodiment of this invention. It is a typical explanatory view of a change of an electron beam orbit when chromatic aberration is included in an electromagnetic lens. It is a theoretical calculation STEM image when using a SrTiO ₃ single crystal having a (001) plane as a sample before and after chromatic aberration correction. It is explanatory drawing of the defocus amount dependence of the contrast with respect to each chromatic aberration coefficient by theoretical calculation. It is explanatory drawing of a STEM image. It is a continuous defocus change STEM image. It is explanatory drawing of the correlation of the full width at half maximum of a normal distribution function, and a chromatic aberration coefficient. It is a flowchart of the chromatic aberration coefficient measuring method of the electromagnetic lens of Example 1 of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Electron gun 12 Electron beam 13 Irradiation lens 14 Spherical aberration correction apparatus 15 Objective lens 16 Sample 17 STEM detector 18 Display unit 19 Control unit 21 Electromagnetic lens 22 Electron beam 23 Imaging surface 31 Focusing electron beam 32 Sample

Claims (6)

A continuous electron beam that passes through a spherical aberration-corrected electromagnetic lens is incident on a single crystal sample with a known crystal structure and lattice constant, and is continuously acquired when a high-resolution scanning transmission electron microscope image with a resolution of 0.3 nm or less is acquired. A step of acquiring a microscopic image in which the amount of defocus of the electromagnetic lens is changed by actual measurement,
A regression curve acquisition step for obtaining a regression curve from an intensity profile obtained by averaging the microscopic image in the scanning direction;
A chromatic aberration coefficient calculating step of calculating a chromatic aberration coefficient of the electromagnetic lens by comparing a chromatic aberration coefficient obtained from a calculation result of a theoretical microscope image obtained by theoretical calculation under the same conditions as the microscope image and a parameter of the regression curve; Method of measuring chromatic aberration coefficient in electromagnetic lens having The method for measuring a chromatic aberration coefficient in an electromagnetic lens according to claim 1, wherein a defocus amount of the electromagnetic lens with respect to the single crystal sample in the step of acquiring the microscopic image by actual measurement is set between -100 nm and 100 nm. When the control unit calibrates the relationship between the defocus amount of the single crystal sample and the current value of the electromagnetic lens, the defocus amount and the current value are calculated using a Ronchigram photographed by the electromagnetic lens before spherical aberration correction. The method for measuring a chromatic aberration coefficient in an electromagnetic lens according to claim 1, wherein the method is calibrated and fed back to the control unit. The electromagnetic lens according to any one of claims 1 to 3, wherein a normal distribution function is used as the regression curve, and a parameter obtained by the regression curve being a normal distribution function is used as a parameter of the regression curve. Chromatic aberration coefficient measurement method. An electron beam generator for generating an electron beam;
A converging lens that converges the electron beam generated by the electron beam generator;
A spherical aberration corrector for correcting the spherical aberration of the convergent lens;
A scanning coil unit that scans a convergent electron beam applied to the sample;
A detector for detecting the diffraction image intensity at the diffraction image surface of the sample;
A display unit for imaging the detection intensity by the detector;
A controller that changes the current value of the convergent lens as the amount of defocus of the sample,
And an arithmetic means for calculating an average intensity profile of the acquired diffraction image;
A scanning transmission electron microscope provided with a chromatic aberration coefficient measurement mechanism for an electromagnetic lens having at least chromatic aberration coefficient calculation means for calculating a chromatic aberration coefficient by comparing the calculation result obtained by the calculation means with a theoretical calculation result. The control unit calibrates the defocus amount of the electromagnetic lens as a change in the current value of the electromagnetic lens; and
The scanning transmission electron microscope according to claim 5 , having a function of automatically changing the current value of the electromagnetic lens continuously and at regular intervals.

Images