JP5413095B2 - Sample analysis apparatus and sample analysis method - Google Patents

Description

  The present invention relates to a sample analysis apparatus and a sample analysis method.

  With the development of the spherical aberration correction device, it has become possible to form a high-current incident electron beam probe, and the analysis method of the electron microscope has been greatly developed. In particular, in high-resolution observation technology using STEM (Scanning Transmission Electron Microscope), an image with atomic resolution is obtained by actively using inelastic scattering, which has been treated as background and noise in conventional microscope images. Has been acquired. A typical example is a method called HAADF (High-Angle Angular Dark-Field) STEM image using a high-angle annular detector. In this method, an image is formed using inelastically scattered electrons due to thermal scattering, and a high resolution image depending on the atomic number is obtained. More recently, the STEM-EELS method, which simultaneously measures EELS (Electron Energy Loss Spectroscopy) while acquiring a STEM image, has begun to attract attention as a method for performing state analysis at atomic resolution. This method obtains an EELS mapping at atomic resolution by storing the EELS spectrum acquired at each incident position and integrating signals in an appropriate energy range for each incident point. In particular, the image obtained by specifying the energy region corresponding to the energy loss due to the inner shell excitation with the energy slit corresponds to the element distribution image, so that it is possible to obtain a composition distribution image at atomic resolution.

  However, since the loss signal due to inner shell excitation is very weak, it is difficult to obtain a STEM-EELS image with high quality atomic resolution.

JP 2003-249186 A

K. Kimoto, T. Asaka, T. Nagai, M. Saito, Y. Matsui, and K. Ishizuka, Nature, 29 November 2007, vol.450, pp.702-704 D. A. Muller, L. Fitting Kourkoutis, M. Murfitt, J. H. Song, H. Y. Hwang, J. Silcox, N. Dellby, and O. L. Krivanek, Science, 22 February 2008, vol.319, pp.1073-1076 L. J. Allen, S. D. Findlay, M. P. Oxley and C. J. Rossouw, Ultramicroscopy, 2003, vol.96, pp.47-63 MP Oxley, M. Varela, TJ Pennycook, K. van Benthem, SD Findlay, AJD'Alfonso, LJ Allen, and SJ Pennycook, Phys. Rev. B, 2007, vol.76, pp.064303-1-064303- 8 Z. L. Wang, Elastic and Inelastic Scattering in Electron Diffraction and Imaging, United States of America, Plenum Press, 1995, pp.167-171 K. Watanabe, T. Yamazaki, I. Hashimoto, and M. Shiojiri, Phys. Rev. B, 2001, vol.64, pp. 115432-1-115432-5

  From a spectrum acquired by EELS, a signal due to plasmon absorption in which energy lower than energy lost due to inner shell excitation is lost can also be detected. Since the signal intensity of this plasmon absorption is much stronger than the signal intensity due to inner shell excitation, an image with high contrast can be obtained. However, since plasmon absorption is known as collective excitation due to interband transition, even if an image by plasmon absorption was acquired at atomic resolution, the meaning represented by the image could not be clearly discussed. This is because there is no method for theoretically calculating the plasmon loss image. Therefore, it is difficult to determine the structure from the intensity of the obtained image.

  In view of the above points, an object of the present invention is to provide a sample analysis apparatus and a sample analysis method capable of easily calculating a plasmon loss image by calculation.

In order to achieve the above object, the following sample analyzer is provided.
The sample analyzer includes an experimental condition including an energy slit set in a plasmon absorption region, an input unit that acquires sample information, a loss function calculation unit that calculates a loss function based on the sample information, and the loss Based on the function and the energy slit set in the plasmon absorption region, an absorption potential calculation unit that calculates two types of absorption potential depending on whether the energy slit is considered, and the two types of absorption potential Based on the total scattered electron intensity calculation unit for calculating the first and second total scattered electron intensities, and a plasmon loss image from the difference between the first total scattered electron intensity and the second total scattered electron intensity. A plasmon loss image generation unit to be generated.

  According to the disclosed sample analysis apparatus and sample analysis method, a plasmon loss image can be easily calculated by calculation.

It is a figure which shows the structure of a sample analyzer. It is a specific hardware configuration example of the sample analyzer. It is a flowchart which shows the flow of an example of a process of a sample analysis method. It is a figure which shows an example of the experimental conditions and sample information of a STEM-EELS method. It is a figure which shows an example of the EELS spectrum detected with an EELS detector. It is a figure which shows an example of a loss function. It is a conceptual diagram of the absorption potential calculated without considering the energy slit. It is a figure which shows the integration range of the loss function in consideration of the energy slit. It is a conceptual diagram of the absorption potential calculated in consideration of the energy slit. It is a conceptual diagram of two types of total scattered electron intensity. A diagram showing a calculation image and experimental images of the plasmon loss image shows (A) the Si (011) surface plasmon loss image, (B) is SrTiO ₃ (strontium titanate) (001) plane plasmon loss image FIG.

Embodiments of a sample analysis apparatus and a sample analysis method of the present invention will be described below with reference to the drawings.
FIG. 1 is a diagram showing a configuration of a sample analyzer.

The sample analysis apparatus 10 according to the present embodiment approximately calculates a high-resolution plasmon loss image by dynamic calculation, which is a theoretical calculation method for a microscopic image.
The sample analysis apparatus 10 includes an input unit 11, a loss function calculation unit 12, an absorption potential calculation unit 13, a total scattered electron intensity calculation unit 14, a plasmon loss image generation unit 15, and a structure analysis unit 16. The input unit 11 includes a sample information input unit 11a and an experimental condition input unit 11b.

  The sample information input unit 11a acquires sample information from, for example, a scanning transmission electron microscope (hereinafter referred to as STEM) 17 main body. The sample information includes information on the crystal structure of the sample and the thickness (film thickness) of the sample.

The STEM 17 includes an EELS detector and can measure EELS simultaneously while acquiring a STEM image.
The experimental condition input unit 11b acquires experimental conditions from the STEM 17 body, for example. The experimental conditions include the acceleration voltage of the incident electron beam, the lens constant, the convergence angle of the incident electron beam, the EELS detector capture angle, the range of the energy slit, and the like in the STEM 17. In the present embodiment, the energy slit is set in a plasmon absorption region that appears in a lower energy region than the loss energy due to inner shell excitation in the EELS spectrum. Plasmon absorption causes energy absorption due to the band structure of the substance. The physical function governing the energy absorption is a function called a loss function obtained from the dielectric function.

  The loss function calculation unit 12 obtains a dielectric function by first-principles band calculation based on information (for example, lattice constant) about the crystal structure of the sample acquired as one of the sample information, and the loss function is calculated from the dielectric function. Is calculated.

  The absorption potential calculation unit 13 calculates two types of absorption potentials based on the loss function obtained by the loss function calculation unit 12 and the energy slit acquired as one of the experimental conditions. The two types of absorption potentials are an absorption potential calculated considering the energy slit and an absorption potential calculated without considering the energy slit. Specifically, the absorption potential considering the energy slit is calculated by removing the range of the energy slit from the integration range of the loss function. On the other hand, the absorption potential that does not consider the energy slit is calculated using the entire energy range of the loss function as an integration range (details will be described later).

The total scattered electron intensity calculation unit 14 calculates each total scattered electron intensity from two types of absorption potentials.
The plasmon loss image generation unit 15 generates a plasmon loss image by calculating a difference between two types of total scattered electron intensities corresponding to two types of absorption potentials at each incident position of an electron beam. Display on the device.

  The structural analysis unit 16 obtains a plasmon loss image (calculated image) calculated by the sample analyzer 10 and a plasmon loss image (experimental image) obtained from the STEM 17 body, and compares the plasmon loss image (experimental image). decide.

Note that the structure analysis unit 16 may be another computer connected to the sample analysis apparatus 10 or the STEM 17.
According to the sample analyzer 10 as described above, a plasmon loss image indicating the atomic structure of the sample can be easily calculated. Moreover, it becomes possible to compare the plasmon loss image obtained by the experiment with the plasmon loss image obtained by the calculation, and the plasmon loss image obtained by the experiment can be easily interpreted. Therefore, the atomic structure of the sample can be easily and accurately analyzed.

Hereinafter, the sample analysis apparatus and the sample analysis method of the present embodiment will be described in more detail and specifically.
FIG. 2 is a specific hardware configuration example of the sample analyzer.

  The sample analysis apparatus 10 shown in FIG. 1 is a computer 20 as shown in FIG. 2, for example. The computer 20 includes a central processing unit (CPU) 21, a read only memory (ROM) 22, a random access memory (RAM) 23, a hard disk drive (HDD) 24, and a graphic processing unit 25. The computer 20 also has an input I / F (Interface) 26, a communication I / F 27, and the like as interfaces. These are connected to each other via a bus 28.

Here, the CPU 21 controls each part of the computer 20 according to programs stored in the ROM 22 and the HDD 24 and various data, and performs the functions of each part shown in FIG.
The ROM 22 stores basic programs and data executed by the CPU 21.

The RAM 23 stores programs being executed by the CPU 21 and data being calculated.
The HDD 24 stores an OS (Operation System) executed by the CPU 21, a program for calculating a plasmon loss image, various application programs, various data, and the like.

  For example, a display 25 a is connected to the graphic processing unit 25. In accordance with a drawing command from the CPU 21, a plasmon loss image generated by calculation or a plasmon loss image obtained by an experiment using the STEM 17 is displayed on the display 25a.

Input devices such as a mouse 26 a and a keyboard 26 b are connected to the input I / F 26, and information input by the user is received and transmitted to the CPU 21 via the bus 28.
The communication I / F 27 is connected to, for example, the STEM 17 or a network 27a such as an in-house LAN (Local Area Network) or WAN (Wide Area Network) to receive experimental conditions, sample information, and an experimental image of a plasmon loss image. And send analysis results.

FIG. 3 is a flowchart showing an exemplary flow of processing of the sample analysis method.
First, in the computer 20 as shown in FIG. 2 which is a sample analyzer, the CPU 21 causes the communication I / F 27 to acquire sample information and experimental conditions from the STEM 17, and stores such information in, for example, the RAM 23 (step S1, S2).

FIG. 4 is a diagram showing an example of experimental conditions and sample information of the STEM-EELS method.
In the STEM-EELS method, an electron beam is converged by a lens 30 and irradiated on a sample 31, and transmitted electrons scattered by the sample 31 are detected by a donut-shaped ADF (Annular Dark-Field) detector 32. Further, EELS due to scattered electrons is detected by the EELS detector 33.

The sample information includes the thickness t of the sample 31, information relating to the crystal structure of the sample 31 (for example, lattice constant), and the like.
Experimental conditions include the incident electron beam convergence angle α, the incident electron beam acceleration voltage, the lens constant, the EELS detector 33 capture angle β, and the energy slit range.

  FIG. 5 is a diagram illustrating an example of an EELS spectrum detected by an EELS detector. The horizontal axis represents energy loss (loss energy) (eV), and the vertical axis represents intensity (au). In FIG. 5, the detected EELS spectrum is classified into a zero-loss spectrum, a low-loss spectrum, and a core-loss spectrum. For the low-loss spectrum and the core-loss spectrum, the intensity is increased 10 times.

  The zero-loss spectrum shows the energy spread of the original electron beam, and the core-loss spectrum is a spectrum in the energy region corresponding to the energy loss due to the inner shell excitation. A low-loss spectrum is a spectrum in an energy region corresponding to plasmon absorption. Plasmon absorption is energy absorption due to the band structure (dielectric characteristics) of the crystal. When a plasmon loss image is obtained by actually using the STEM 17, as shown in FIG. 5, an energy slit is inserted into this low-loss spectrum, and a two-dimensional image is calculated based on the spectrum in that region. For example, the energy slit is set in a range of about 5 eV with the peak of the spectrum as the center. An energy slit may be set in a range narrower than 5 eV for a peak that is easily understood, and a range wider than 5 eV for a peak that is difficult to understand.

Also in the sample analysis method of the present embodiment, in order to add the effect of inserting this energy slit, the CPU 21 acquires the range of the energy slit which is an experimental condition.
In FIG. 3, step S1 for obtaining sample information and step S2 for obtaining experimental conditions may be switched in order. The computer 20 may acquire sample information and experimental conditions from another computer that sets experimental conditions and the like in the STEM 17 or may be acquired from the user via the input I / F 26.

  When the sample information and the experimental conditions are acquired, the CPU 21 calculates the loss function using the first principle band calculation based on the information on the crystal structure of the sample (step S3). In the first-principles band calculation, for example, an approximation such as LDA (Local Density Approximation) or GGA (Generalized Gradient Approximation) is used, and the dielectric function is calculated based on the crystal structure information such as the lattice constant. And a loss function is calculated | required from a dielectric function.

FIG. 6 is a diagram illustrating an example of the loss function. The horizontal axis represents energy (eV), and the vertical axis represents the magnitude of the loss function.
Here, the loss function of Si (silicon) calculated by the first principle band calculation is illustrated. The loss function of Si shows the maximum value near 17 eV.

Next, the CPU 21 calculates two types of absorption potentials based on the calculated loss function and experimental conditions (step S4).
One of the two types of absorption potentials is an absorption potential calculated by integrating in the entire energy region of the loss function without considering the energy slit.

  FIG. 7 is a conceptual diagram of the absorption potential calculated without considering the energy slit. The horizontal axis is energy (eV), and the vertical axis is intensity (au). This absorption potential includes the total energy band absorbed by the plasmon loss, and is calculated by the following equation (for details, see pages 167 to 171 of Non-Patent Document 5).

Here, Im [−1 / ε (E)] is the loss function calculated in the process of step S3. Im [−1 / ε (E)] represents an imaginary part of −1 / ε (E).
Another one of the two types of absorption potential is an absorption potential calculated in consideration of the energy slit. Specifically, the CPU 21 calculates the absorption potential by removing the energy slit range from the integration range.

FIG. 8 is a diagram illustrating an integration range of the loss function in consideration of the energy slit. The horizontal axis represents energy (eV), and the vertical axis represents the magnitude of the loss function. Here, the integration range in the loss function of Si shown in FIG. 6 is shown. The shaded area is the integration range, and the energy slit ranges E _{1 to} E ₂ are not integrated.

  The absorption potential considering the energy slit is calculated by the following equation.

In the expression (2), unlike the expression (1), the range from E ₁ to E ₂ is excluded from the integration range.
FIG. 9 is a conceptual diagram of the absorption potential calculated in consideration of the energy slit. The horizontal axis is energy (eV), and the vertical axis is intensity (au). This absorption potential is obtained by reducing the effect of plasmon absorption.

  When the calculation of the two types of absorption potentials as described above is completed, the CPU 21 calculates two types of total scattered electron intensities based on the two types of absorption potentials, experimental conditions, and sample information (step S5). The total scattered electron intensity is obtained by the following equation (refer to Non-Patent Document 6 for details).

Here, when obtaining the transmission coefficient T _g of the transmitted electron beam, the absorption potential Va obtained in the processing in step S4, Vb is used, the absorption potential Va, different transmission coefficient T _g according to Vb obtained. As a result, two types of total scattered electron intensities are obtained.

  FIG. 10 is a conceptual diagram of two types of total scattered electron intensities. The horizontal axis indicates the thickness (nm) of the sample, and the vertical axis indicates the intensity (au). The total scattered electron intensity obtained based on the absorption potential Va is indicated by Ia, and the total scattered electron intensity obtained based on the absorption potential Vb is indicated by Ib.

  Although the total scattered electron intensity is attenuated by the thickness of the sample, there is a difference in intensity between the two types of total scattered electron intensity Ia and Ib. This difference in intensity is due to whether plasmon absorption is taken into account. In other words, this difference can be regarded as the intensity of the plasmon-absorbed electron.

The CPU 21 varies R ₀ (the center of the position of the incident electron beam) in Equation (3) and plots the difference between the two types of total scattered electron intensities at each incident point of the electron beam, thereby obtaining a plasmon loss image. Is generated (step S6).

By the sample analysis method as described above, a plasmon loss image showing the atomic structure of the sample can be easily obtained by calculation.
Next, the CPU 21 performs structural analysis processing based on the plasmon loss image obtained by calculation and the plasmon loss image obtained by the experiment using the STEM 17 (step S7).

  Specifically, the CPU 21 captures the plasmon loss image obtained by actual measurement by the STEM 17 via the communication I / F 27 into the computer 20 and displays it on the display 25a together with the plasmon loss image obtained by calculation.

FIG. 11 is a diagram showing an experimental image and a calculated image of a plasmon loss image. (A) is a plasmon loss image of the Si (011) plane, and (B) is a plasmon of SrTiO ₃ (strontium titanate) (001) plane. Shows Ross image. 11A and 11B, the left shows an experimental image, and the right shows a calculated image.

  As an experimental condition, in the example of FIG. 11A, the energy slit range was 15 to 18 eV, and in the example of FIG. 11B, the energy slit range was 28 to 32 eV. In both cases, the take-in angle of the EELS detector 33 was 20 mrad. This angle is about the same as the convergence angle of the incident electron beam.

  As shown in FIGS. 11A and 11B, in both cases, the calculated image well reproduces the characteristics of the experimental image. In addition, it is confirmed that the atomic position is detected as a weak intensity point (black point) in the plasmon loss image by setting the capture angle of the EELS detector 33 to be approximately the same as the convergence angle of the incident electron beam. It was done. The higher the atomic number, the weaker the intensity and the clearer it appears as a black dot. For example, the CPU 21 compares the calculated image and the experimental image, and if the position of the black dot matches the calculated image and the experimental image, the CPU 21 specifies the position as the atomic position.

  As described above, according to the sample analysis method of the present embodiment, the plasmon loss image obtained by measurement and the plasmon loss image obtained by calculation can be compared, and the plasmon loss obtained by measurement can be compared. Interpretation of the image becomes easy.

Therefore, the atomic structure of the sample can be easily and accurately analyzed.
As mentioned above, although one viewpoint of the sample analyzer and sample analysis method of this invention was demonstrated based on embodiment, these are only examples and are not limited to said description.

DESCRIPTION OF SYMBOLS 10 Sample analyzer 11 Input part 11a Sample information input part 11b Experimental condition input part 12 Loss function calculation part 13 Absorption potential calculation part 14 Total scattered electron intensity calculation part 15 Plasmon loss image generation part 16 Structure analysis part 17 STEM

Claims (5)

Experimental conditions including an energy slit set in the plasmon absorption region, an input unit for acquiring sample information,
A loss function calculator that calculates a loss function based on the sample information;
Based on the loss function and the energy slit set in the plasmon absorption region, an absorption potential calculation unit that calculates two types of absorption potentials depending on whether the energy slit is considered,
Based on the two types of absorption potentials, a total scattered electron intensity calculation unit for calculating first and second total scattered electron intensity,
A plasmon loss image generation unit that generates a plasmon loss image from a difference between the first total scattered electron intensity and the second total scattered electron intensity;
A sample analyzing apparatus comprising:   The absorption potential calculation unit calculates a first absorption potential with an integration range outside the energy slit range, and calculates a second absorption potential with an energy range including the energy slit range as an integration range. The sample analyzer according to claim 1, wherein the sample analyzer is characterized in that:   A structural analysis unit that acquires an experimental image of the plasmon loss image, compares the calculated image of the plasmon loss image generated by the plasmon loss image generation unit with the acquired experimental image, and determines an atomic position. The sample analysis apparatus according to claim 1, wherein:   The sample analysis according to any one of claims 1 to 3, wherein, among the experimental conditions, the capture angle of the detector for electron energy loss spectroscopy is substantially the same as the convergence angle of the incident electron beam. apparatus. Obtain the experimental conditions including the energy slit set in the plasmon absorption region and the sample information,
Calculate a loss function based on the sample information,
Based on the loss function and the energy slit set in the plasmon absorption region, two types of absorption potentials are calculated depending on whether or not the energy slit is considered,
Based on the two kinds of absorption potentials, the first and second total scattered electron intensities are calculated,
A sample analysis method, wherein a plasmon loss image is generated from a difference between the first total scattered electron intensity and the second total scattered electron intensity.

Images