JP2011066005A - Charged particle beam device, and sample preparation method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To carry out with a high throughput both of a thin-film working of high precision by ion beam irradiation and a STEM observation of high resolution by electron beam irradiation of a sample, without moving the sample. <P>SOLUTION: An irradiation axis of an FIB irradiation system 16 and an irradiation axis of a STEM observation electron beam irradiation system 5 are made almost orthogonal, and the sample 7 is arranged at the intersection position and the FIB cross-section working face of the sample is made a thin-film face of the STEM observation sample. Transmission-scattering beam detection means 9a, 9b are arranged at the rear part of the sample viewed from electron beam irradiation direction on the electron beam irradiation axis. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a charged particle beam apparatus capable of observing, for example, a fine sample taken out from a substrate of a semiconductor device while being finely processed, and a sample preparation method using the charged particle beam apparatus.

  A microfabrication technique using a focused ion beam (FIB) is described in, for example, the republished patent WO99 / 05506. By irradiating the sample with FIB, microfabrication utilizing the sputtering phenomenon can be performed, so that a fine sample can be excavated from a substrate of a semiconductor device, for example. Further, if a deposition gas is introduced near the FIB irradiation portion of the sample and FIB irradiation is performed in the gas atmosphere, a deposition film can be formed by an ion beam assisted deposition phenomenon. This film formation can be used for a kind of fine adhesion. By installing a probe for pulling up a fine sample in the FIB sample chamber equipped with a deposition gas supply source, the fine sample dug by FIB processing is fixed to the probe by FIB deposition film formation and separated from the sample substrate. Can be raised. The pulled-up fine sample can be fixed to a fine sample holding part previously placed near the sample substrate in the same sample chamber. The fine sample holder has a shape that can be easily mounted on a sample holder of a transmission electron microscope (TEM) or a scanning transmission electron microscope (STEM) from the viewpoint of operability.

  Techniques relating to the combination of the FIB apparatus and the STEM are disclosed in Japanese Patent Application Laid-Open Nos. 2004-228076 and 2002-29874. Japanese Patent Application Laid-Open No. 2004-228076 discloses that a sample for STEM observation produced by FIB processing is placed at the intersection of an ion beam axis and an electron beam axis, so that additional processing by FIB and STEM observation can be performed. The thin film surface of the STEM sample needs to be placed almost parallel to the additional processing by FIB and almost perpendicular to the STEM observation. However, in this technique, the ion beam axis and the electron beam axis intersect at an acute angle (about 45 degrees in FIG. 5), and the STEM sample can be rotated around the rotation axis perpendicular to both axes between the FIB additional processing and the STEM observation. Necessary. Also in Japanese Patent Application Laid-Open No. 2002-29874, the ion beam axis and the electron beam axis are an example of an acute angle (about 50 degrees in FIG. 1).

  On the other hand, a technique related to TEM sample preparation is disclosed in Japanese Patent Laid-Open No. 6-231720. In order to investigate the film thickness of a TEM sample processed into a thin film by FIB, an electron beam is irradiated from a direction perpendicular to the cross section of the sample, and the intensity ratio between the electron beam irradiation intensity and the electron beam transmitted through the sample is detected. Yes. Here, the transmitted electron beam is not detected by distinguishing between a transmitted beam having a very small scattering angle as a STEM signal and a scattered beam that has been transmitted and scattered, and is not a sample film thickness monitor from the viewpoint of STEM observation. . The same kind of technique is also disclosed in Japanese Patent Laid-Open No. 8-5528. In JP-A-6-231719, an FIB irradiation system is arranged in a sample chamber of a TEM apparatus in a direction perpendicular to the electron beam axis of the TEM, and a TEM sample produced by FIB processing is not taken out into the atmosphere as it is. It is characterized by being observable, and is disclosed as an apparatus that solves poor throughput such as during FIB additional processing. There, a TEM image monitor is also disclosed as a sample film thickness control.

  JP-A-7-92062 discloses an example of in-situ TEM observation of a sample during FIB processing by attaching the FIB irradiation system axis in an oblique direction (<90 degrees) with respect to the electron beam irradiation system axis of the TEM apparatus. Yes. However, since the FIB is obliquely incident on the sample surface, there is a drawback in that the beam damage is deeply formed as compared to the cross-sectional thin film sample processed and fabricated almost parallel to the FIB incident axis.

  On the other hand, the STEM single device is described in, for example, Japanese Patent Application Laid-Open No. 2000-21346. The beam scanning images used for the luminance signal by detecting the transmitted electrons and scattered electrons from the sample are called a bright field image and a dark field image, respectively. The bright field image contrast reflects the absorption, diffraction, and phase shift of the irradiated electrons in the sample. On the other hand, the dark field image has a characteristic that only the diffracted electron reflected in a specific angular direction is used as a luminance signal, and the size of the crystal grains clearly appears. Both the bright-field image and the dark-field image are images (STEM images) formed by the beam that has passed through the sample, and are very effective for analyzing the structure in the thin film.

  In addition, the energy of the irradiation electron beam of the STEM apparatus has been as high as about 200 keV as in the case of TEM, but recently it seems to have been reported in the paper Journal Electron Microscopy 51 (1): 53-57 (2002). In addition, STEM observation has been performed in the same energy region as that of a conventional SEM of 30 keV. When the irradiation energy of the electron beam is reduced, the transmission capability of the thin film sample is also reduced. Therefore, in a low energy STEM apparatus, a thin film sample manufactured by FIB processing is generally required to have high precision film thickness control.

Republished patent WO99 / 05506 JP 2004-228076 A JP 2002-29874 A JP-A-6-231720 JP-A-8-5528 JP-A-6-231719 JP-A-7-92062 JP 2000-21346 A JP 2003-142021 A

Journal Electron Microscopy 51 (1): 53-57 (2002)

  However, many FIB processing and STEM observations have been performed with separate devices so far, and the thin film sample for STEM processed and manufactured with the FIB device must be taken out of the FIB device and then put into the STEM device for observation. was there. Therefore, the need for thin film processing while further specifying the observation location by repeating STEM observation and FIB additional processing has not been sufficiently met from the viewpoint of throughput. Recently, an apparatus in which FIB processing and STEM observation are integrated has also been put out, but there remains an operation for rotating the sample during FIB processing and STEM observation and aligning the sample thin film surface in each direction.

  From the viewpoint of STEM image observation, it is important to ensure sufficient image signal intensity and high image contrast, but they are usually incompatible (rather, contradictory). Especially in dark field images, the optimum sample thickness differs depending on the sample material (atomic number), the observation site, and the uniformity in the thickness direction at the sample interface of the observation target. Is difficult. Therefore, the sample is gradually thinned from the thicker direction to the optimum thickness while monitoring the STEM image during the processing.

  The first problem to be solved by the present invention is the omission (or minimization) of the sample rotation operation between the FIB processing and the STEM observation, and the sample using the STEM image monitor during the processing. An object of the present invention is to provide an apparatus capable of realizing high throughput in one sample chamber from FIB processing to STEM observation by simplifying the thickness optimization operation.

  On the other hand, in STEM observation with low energy (about 30 keV), the image resolution is lowered by the amount of the beam diameter becoming thicker compared with observation with high energy (about 200 keV). Thus, there are apparatus users who wish to increase the image resolution of STEM observation even at the expense of some simplicity of operation and throughput. A second problem is to provide an apparatus that emphasizes such STEM image resolution.

  In order to solve the first problem, the ion beam irradiation system, the electron beam irradiation system, and the transmitted / scattered beam detection means are provided in the sample so that both FIB processing and STEM observation can be performed without moving the sample as much as possible. It is to be placed around. That is, the irradiation axis of the FIB irradiation system and the irradiation axis of the electron beam irradiation system for STEM observation are almost orthogonal to each other, the sample is arranged at the intersection, and the processed surface of the FIB cross section of the sample is taken as the thin film surface of the sample for STEM observation. It is. Further, the transmitted / scattered beam detecting means is arranged behind the sample as viewed from the electron beam irradiation direction on the electron beam irradiation axis. With this arrangement, FIB processing and STEM observation can be realized with little movement of the sample.

  Next, means for solving the second problem will be described. The first is the simplification of the work of omitting (or minimizing) the work of rotating the sample between FIB processing and STEM observation, and the work of optimizing the sample thickness using the STEM image monitor during processing. From this point of view, the irradiation axis of the FIB irradiation system and the irradiation axis of the electron beam irradiation system for STEM observation are almost orthogonal. On the other hand, from the viewpoints of beam bundles (high processing speed and high processing position accuracy) in FIB processing and high image resolution in STEM observation, an electron beam irradiation system or an ion beam irradiation system objective lens can be applied to the sample as much as possible. It is required to be placed in a close position. However, there is a limit to the distance that can be approached to the sample due to the spatial size of the objective lens of both irradiation systems. Crossing the irradiation axis of the FIB irradiation system and the irradiation axis of the electron beam irradiation system for STEM observation at an angle exceeding a right angle under the condition that the image resolution of STEM observation is increased even at the expense of some simplicity of operation and throughput. It is effective to make it. According to this arrangement, the objective lens of the electron beam irradiation system can be brought closer to the sample than when the irradiation axis of the FIB irradiation system and the irradiation axis of the electron beam irradiation system for STEM observation are arranged substantially at right angles.

  When the objective lens of the electron beam irradiation system is a magnetic field type lens, the objective lens is effectively brought close to the sample by leaking the magnetic field to the sample side. This is also a means for solving the high performance problem of STEM observation. When the FIB is composed of isotope ions, the leakage magnetic field causes separation at the beam irradiation point on the sample in order to separate the ion beam trajectory in the magnetic field. To solve this problem, it is only necessary to form a compensation magnetic field so that the separated beam is again focused on one point on the sample.

  According to the present invention, a thin film sample is produced with higher fine processing capability in FIB processing, while STEM observation with higher resolution can be performed in STEM observation. Further, the film thickness can be controlled with high accuracy in the preparation of a thin film sample, and the processing from FIB processing to STEM observation can be realized in a single sample chamber with high throughput in time series or simultaneously.

The figure which shows schematic structure of the 1st Example of the charged particle beam apparatus by this invention. FIG. 2 is a cross-sectional view taken along line A-A ′ of FIG. 1. The schematic diagram explaining arrangement | positioning of the secondary electron detection surface in the cross section containing the linear axis drawn from the sample to the secondary electron detection surface and an electron beam irradiation axis. The figure explaining the whole apparatus structure of the beam irradiation system of this invention, a beam control part, and an image display means. (A) is an example of a graph in which the brightness and contrast in a bright field image are displayed with respect to the shaving thickness (arbitrary unit) when the sample is finished by FIB, and (b) is a region including the interfaces A and B in the dark field image. The figure which shows the example of the graph displayed with respect to the shaving thickness (arbitrary unit) of the contrast in. The figure which shows the example which showed the bright field image, the dark field image, the SIM image, and the display graph in the display image frame of an image display means. The figure which shows schematic structure of the 2nd Example of the charged particle beam apparatus by this invention. The figure explaining the crossing angle of an electron beam irradiation axis and an ion beam irradiation axis. The combination explanatory drawing of the electron beam irradiation system and the ion beam irradiation system which made the objective lens the magnetic field type lens, and made the magnetic field leak to the sample side. The magnetic flux distribution example figure on the ion beam irradiation axis | shaft of the magnetic field leaked from the objective lens of the electron beam irradiation system. Explanatory drawing of the ion beam irradiation system incorporating the correction magnetic field generation part. Schematic of a correction magnetic field generation part.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a diagram showing a schematic configuration of a charged particle beam apparatus according to a first embodiment of the present invention, and FIG. 2 is a cross-sectional view taken along line A-A ′ of FIG.

  The electron beam irradiation system 5 and the ion beam irradiation system 16 are attached to the sample chamber casing 11 so as to irradiate the sample 7. The electron beam irradiation system 5 includes a focusing lens 2 for focusing electrons emitted from the electron gun 1, an electron beam deflector 3, and an objective lens 4. The electron beam irradiation system 5 is narrowed down and scanned as an electron beam 6 on a sample 7. Has a function to irradiate. On the other hand, the ion beam irradiation system 16 includes a focusing lens 13 that focuses ions emitted from the ion gun 12, an ion beam deflector 14, and an objective lens 15, and is narrowed as a focused ion beam (FIB) 17 on the sample 7. A diaphragm and a function of irradiating while scanning within a specific range at a specific location. In the embodiment of FIG. 1, the irradiation axes of the electron beam 6 and the ion beam 17 on the sample 7 substantially intersect, and the intersecting angle is substantially a right angle (= 90 degrees ± 1 degree). The sample is placed at the intersection 30 and thin film processing by ion beam irradiation and STEM observation by electron beam irradiation can be performed with good operability without moving the sample.

  A part of the sample is thinned by FIB processing, and the electron beam 6 is irradiated and scanned on the thin piece portion 7a. The transmitted beam 8a and scattered beam 8b that have passed through the thin film portion 7a are detected by a transmitted beam detector 9a and a scattered beam detector 9b, respectively. Other detectors include secondary electron detection means 20 and backscattered electron detection means 25. The secondary electron detection means 20 detects the secondary electrons 19 emitted from the sample 7 during FIB processing. In FIG. 1, the secondary electron detection means 20 is actually located behind the ion beam irradiation system 16 and cannot be seen, and therefore is drawn just above the ion beam irradiation system 16 for convenience. Actually, as shown in FIG. 2 showing the AA ′ cross section (plane perpendicular to the irradiation axis of the electron beam 6 and including the irradiation axis of the ion beam 17) in FIG. 'Located in the cross section.

  When a straight line drawn from the sample 7 to the secondary electron detection means 20 is an axis 50, this axis is in the A-A 'cross section. FIG. 3 is a cross-sectional view including the axis 50 and the irradiation axis of the electron beam 6. The center of the secondary electron entrance surface 21 of the secondary electron detection means 20 is on the axis 50, that is, the secondary electron entrance surface 21 is substantially plane symmetric with respect to the plane including the axis 50 and the irradiation axis of the ion beam 17. Placed in position. Thereby, the secondary electrons 19 emitted from the front and back surfaces of the thin film sample surface 7a by FIB can be collected and detected with substantially the same yield. The scanned image is called a scanning ion microscope image (SIM image). The sample holder 22 has a holder axis 23 that is perpendicular to both the electron beam and ion beam irradiation axes and passes through the intersection 30 of the both irradiation axes. The sample holder 22 is held by a sample stage 26 for finely moving the sample including an inclination, and the sample stage 26 is fixed to the sample chamber housing 11.

  So far, the case where there is one secondary electron detection surface has been described. However, the number of secondary electron detection surfaces may be two as shown by the symbols 21a and 21b in the broken line rectangle of FIG. 3 in order to further increase the secondary electron collection efficiency. Even in such a plurality of arrangements, it is important that the secondary electron entrance surface is arranged at a plane symmetrical position with respect to the plane including the axis 50 and the irradiation axis of the ion beam 17 as described above. The secondary electron detection means 20 is also used when detecting secondary electrons 19 emitted from the sample 7 during electron beam irradiation. In particular, when two (or a plurality) of secondary electron entrance surfaces 21 are arranged almost symmetrically as shown by broken lines in FIG. 3, secondary electrons generated from the front and back surfaces of the sample thin film are separated and detected. Is also possible. When the electron beam is radiated, reflected electrons 24 reflected backward are generated from the sample 7 in addition to the secondary electrons. Therefore, a reflected electron detection means 25 for detecting the reflected electrons 24 is also provided (see FIG. 1). ). Any detection signal can be used as a luminance signal of a scanned image drawn in synchronization with the beam scanning signal.

  FIG. 4 is a diagram for explaining the overall configuration of the beam irradiation system, beam controller, and image display means of the charged particle beam apparatus of the present invention. In FIG. 4, detection signals from all the detection means are taken into the beam control means 40, and beam control of the electron beam irradiation means 5 and the ion beam irradiation means 16 is performed from this beam control means 40. The beam control means 40 includes not only a voltage current source for supplying and supplying current to both beam irradiation means, but also a computer, in which software processing such as voltage current control and signal processing from each detector is also performed. ing. In addition, an image display means 41 for the purpose of these control and processing parameter input / output displays and display of various scanned images is electrically connected to the beam control means 40. Further, for various scanned images, the image average brightness, contrast, signal noise ratio (S / N ratio), and image resolution are evaluated numerically, and numerical display or graph display is performed on the display screen of the image display means 41. Can do. In general, as the thin film sample becomes thinner, the amount of transmitted electron beams also increases, so that the brightness increases and the image resolution increases (the resolution numerical value decreases).

  FIG. 5A is an example of a graph in which the brightness and contrast in a bright field image of STEM are displayed with respect to the shaving thickness (arbitrary unit) when the sample is finished by FIB. FIG. 5B is an example of a graph displayed with respect to the thickness of the contrast (arbitrary unit) in the region including the interfaces A and B in the dark field image of the STEM. The brightness and contrast were values obtained by dividing the average brightness and the 10% -90% brightness range of the pixel brightness histogram in the coarse scan image (512 × 512 pixels) by the average brightness, respectively. Luminance and contrast are calculated in real time at predetermined timings from STEM images acquired during processing, and the results are sequentially added to the graph and plotted.

  In the bright field image, as the sample is made thinner, both brightness and contrast increase as shown in FIG. 5 (a), but the contrast saturates halfway and tends to decrease. The reason for the tendency of saturation and decrease from the middle of contrast is not well understood. In the contrast of the dark field image shown in FIG. 5B, it can be seen that the amount of shaving of the sample that gives the maximum contrast differs depending on the interface regions A and B to be observed. In other words, when performing high-contrast observation of a dark field image in a specific interface region, the optimum sample thickness cannot be determined by monitoring the total beam intensity transmitted through the sample. This can be determined by acquiring a field image and monitoring its contrast. After reaching the optimal sample thickness, the desired high definition STEM images (1024 × 1024, 2048 × 2048, 4096 × 4096 pixels) were acquired.

  In addition to the brightness and contrast of the scanned image, graph display (or numerical display) of the SN ratio and image resolution is also possible. In the display graph, in order to make it easy to see the change in the curve, a mode for displaying the differential curve with respect to the cutting amount is prepared. In FIGS. 5A and 5B, in addition to plotting the brightness and contrast against the shaving amount, the differential amount (arbitrary unit) is also plotted on the right vertical y-axis display. The intersection of the differential plot curve with the y = 0 line corresponds to the maximum value of the original plot curve before differentiation, and this differential plot curve is very effective in grasping the state of arrival of the maximum value. In the graph plot display, it is possible to select whether the original plot or the superimposition of the original plot and the differential plot.

Hereinafter, the calculation method used in the present embodiment will be described with respect to luminance, contrast, SN ratio, and image resolution. A method for evaluating these from images has not yet been established, and research and development is ongoing. First, the luminance is a value obtained by subtracting the origin (the pixel intensity average value when the beam irradiation is stopped) from the average value of the pixel intensity in a specific area in the image designated by the image observer. The contrast is an intensity difference between a maximum of 90% and a minimum of 10% in the histogram of pixel intensity in a specific area. For the S / N ratio and the image resolution, the image evaluation method disclosed in Japanese Patent Laid-Open No. 2003-142021 was used. That is, for each pixel position in the image, a local xy region 3x3 centered on that position is cut out, and the quadratic curved surface expressed by the following equation is fitted by using the pixel intensity there as z (with the coefficients a to f Is determined.)
z = ax ² + by ² + cxy + dx + ey + f (from a to f: coefficient)

  When there is a lot of image noise, this fitting shift increases. The average of this fitting shift in the entire specific region was defined as N of the SN ratio, the luminance described above was defined as S, and the ratio of S to N (= S / N) was defined as the SN ratio. For the image resolution, the reciprocal of the gradient at the center of the quadratic surface pasted on the local xy region 3x3 is defined as a proportional amount of the local image resolution, and the weighted average of the proportional amount in the entire specific region is defined as the image resolution (arbitrary Unit). The absolute value of the gradient was adopted as the weight.

  The graph display such as brightness, contrast, SN ratio and image resolution is while actually observing the STEM image, and it is determined whether or not the FIB processing is continued or stopped, or whether additional processing is necessary, or In particular, it was very effective to make a determination by paying attention to certain specific image evaluation parameters (such as luminance, contrast, SN ratio, and image resolution). In particular, from the viewpoint of automatic processing, when a thin film sample is manufactured by FIB rough processing, threshold values of desired STEM image characteristics (luminance, contrast, SN ratio, image resolution, etc.) are registered in advance as rough end point detection signals. By doing so, automatic processing became possible. A function for registering a threshold value for brightness, contrast, SN ratio, or image resolution, a function for determining whether or not the characteristics of the STEM image have reached a pre-registered threshold value, and the characteristics of the STEM image have reached a pre-registered threshold value When the determination is made, the beam control means 40 can have a function of stopping the ion beam irradiation and ending the rough machining.

  FIG. 6 shows an arrangement example in which a bright field image, a dark field image, a SIM image, and the above display graph are displayed in the display image frame of the image display means 41.

  The usage of the scanned image will be summarized below. The SIM image is used for setting a FIB processing position, a processing monitor during processing, confirmation of processing end, observation of a processing cross section, and the like. On the other hand, the signals obtained by electron beam irradiation include secondary electrons, reflected electrons, transmitted electrons, and scattered electrons, and their scanned images are secondary electron images, reflected electron images, bright field images, and dark electrons, respectively. This is called a field image, and the previous two images are classified as SEM images, and the remaining two images are classified as STEM images. The STEM image is an image formed by the beam that has passed through the sample, and is very effective for analyzing the structure in the thin film. In addition, since the ion irradiation axis and the electron beam irradiation axis intersect at 90 degrees, it is possible to observe an SEM image of the sample during FIB processing from the lateral direction. In a thin film sample manufactured by SEM cross-section processing, the top portion may be damaged by etching at the FIB beam skirt. This SEM image is very effective for monitoring observation of the etching damage. Further, there is no need to move the sample by FIB processing and STEM observation. The combined observation of the SIM image, SEM image, and STEM image is very important from the viewpoint of operability, throughput, and quality of the observed image, particularly when further detailed analysis locations are tracked by FIB processing while viewing the observed image. It was confirmed that there is a big effect. This STEM image (or SEM image) observation and FIB processing can be performed in time series or simultaneously. This makes it possible for the operator to interrupt and resume the FIB processing at any time while monitoring the scanned image, so that high-precision processing of a specific part of the sample and high-resolution STEM observation can be realized reliably and with high throughput. .

  FIG. 7 is a diagram showing a schematic configuration of a second embodiment of the charged particle beam apparatus according to the present invention, and FIG. 8 is a diagram for explaining the crossing angle between the electron beam irradiation axis and the ion beam irradiation axis. The charged particle beam apparatus of the present embodiment is an example in which the crossing angle between the electron beam irradiation axis and the ion beam irradiation axis exceeds 90 degrees.

  In this embodiment, the crossing angle of the irradiation axes of the electron beam 6 and the ion beam 17 to the sample 7 is 110 degrees. As a result, the objective lens 15 of the ion beam irradiation system 16 can be brought closer to the sample by about 3 mm than in the case of Example 1 having an intersection angle of 90 degrees, and the FIB diameter is reduced by about 15% in terms of the current value with the same processing beam performance. did it. This corresponds to an approximately 15% improvement in FIB machining position accuracy under the same machining speed conditions. However, it is necessary to move the sample 7 by an inclination angle of the sample inclination angle difference β (see FIG. 8) between FIB processing and STEM observation. The sample holder 22 is irradiated with both an electron beam and an ion beam in the same manner as in FIG. 2 so that the sample movement between the FIB processing and the STEM observation can be performed only by the inclination angle of the sample inclination angle difference β. It is taken on an axis perpendicular to the axis and passing through the intersection 30 of the two irradiation axes. The sample holder 22 is held by a sample stage 26 for finely moving the sample including an inclination, and the sample stage 26 is fixed to the sample chamber housing 11.

  In order to make the objective lens of the electron beam irradiation system and the objective lens of the ion beam irradiation system closer to the sample, in order to maximize the space near the sample, the crossing angle of both beam irradiation axes should be 180 degrees. Is good. However, since the transmitted / scattered beam detecting means 9a and 9b are arranged in the direction of 180 degrees from the electron beam irradiation axis, they cannot be taken at 180 degrees. Specifically, the cone angle θeb and the irradiation system diameter on the objective lens end side of the electron beam irradiation system 5, the cone angle θib and the irradiation system diameter on the objective lens end side of the ion beam irradiation system 16, and the transmitted / scattered beam detection means The crossing angle of both beam irradiation axes is set between 90 degrees and 180 degrees by balancing the increase in the movement angle β and the beam performance improvement while considering the size and the like.

  When the objective lens of the electron beam irradiation system is a magnetic lens, the objective lens is effectively brought close to the sample by leaking the magnetic field to the sample side, so that the performance of STEM observation is improved. The present embodiment relates to this leakage magnetic field type objective lens and will be described with reference to FIG.

The objective lens 4 is a leakage magnetic field type and includes a magnetic pole piece 4a and an excitation coil 4b. The magnetic field is leaked to the sample side, that is, the intersection 30 side of the irradiation axis of the electron beam 6 and the irradiation axis of the ion beam 17 to form a magnetic flux density distribution 35 on the irradiation axis of the electron beam 6. The leaked magnetic field also exists on the irradiation axis of the ion beam 17. An example of the magnetic flux distribution on the irradiation axis of the ion beam 17 is shown in FIG. Gallium (Ga) is usually used as the ion species of the ion beam for FIB processing. Since Ga has isotopes Ga ⁶⁹ and Ga ⁷¹ , the magnetic flux distribution on the irradiation axis of the ion beam 17 separates the beam spot on the sample and effectively increases the beam diameter in the separation direction. The increase in the beam diameter deteriorates the processing position accuracy, and thus countermeasures are required. The following measures were taken against this.

  When the magnetic field distribution of the magnetic field type objective lens in the electron beam irradiation system is analyzed in detail, for example, a strong magnetic field downward (or upward) exists in a narrow region near the electron beam irradiation axis, and upward ( There is a weak magnetic field (or downward). The ion beam irradiation axis penetrates the magnetic field space in both directions. In order to solve the ion beam spot separation problem, there is a region on the optical axis of the ion beam in which the magnetic field components perpendicular to the optical axis are opposite to each other. If the ion beam path from the ion gun to the sample is different for each isotope, the beam spot position on the sample is matched if the ion beam path does not exist. Specifically, as a result of bringing a correction magnetic field generation unit on the optical axis of the ion beam irradiation system and combining the leakage magnetic field and this correction magnetic field, the beam spot position on the sample coincides with the case where no magnetic field exists. Thus, the correction magnetic field is set.

  FIG. 11 shows an embodiment of an ion beam irradiation system in which a correction magnetic field generator 27 is brought on the optical axis. 12A and 12B are schematic views of the correction magnetic field generation unit, where FIG. 12A is a cross-sectional view, and FIG. 12B is a top view. The correction magnetic field generation unit 27 includes a pair of opposed coils 61 and a permalloy magnetic path 62. The magnetic path 62 serves as a magnetic shield for efficiently generating a correction magnetic field on the ion beam optical axis and at the same time suppressing a leakage magnetic field from the coil 61 to the outside. Here, the material of the magnetic path is not limited to permalloy, and any material may be used as long as it is a magnetic material having a large magnetic permeability and a small coercive force, such as pure iron or permender. As the permeability increases, the leakage magnetic field to the outside can be reduced. The correction magnetic field generator 27 has a beam path in the center, and the beam path is arranged so as to coincide with the irradiation axis of the ion beam 17. The beam path of the correction magnetic field generation unit 27 has a diameter through which all the ion beams that have passed through the focusing lens 13 pass.

  In this figure, the direction of the correction magnetic field 63 generated on the ion beam optical axis is parallel to the paper surface and perpendicular to the optical axis. In this way, a Lorentz force acts on the ion beam in a direction perpendicular to the paper surface and the optical axis, and this direction is parallel to the Lorentz force generated by the magnetic field from the objective lens 4 of the SEM, and the direction of the correction magnetic field and By appropriately adjusting the size, the irradiation spot position of the ion beam on the sample 7 can be completely matched with the position in the absence of a magnetic field.

  Here, the beam spot does not necessarily have a circular shape. Even when the beam cross section is elliptical or linear due to the influence of astigmatism or the like, the discussion here is equally valid. In the present invention, the term “beam spot” includes all shapes. The correction magnetic field generation unit may be located anywhere on the ion beam optical axis, but it is preferable to keep the correction magnetic field generation unit away from the sample 7 so as not to disturb the magnetic field of the SEM objective lens 4. Further, although the correction magnetic field generation unit 27 of FIG. 11 shows a case where a pair of coils is used, two or more pairs of coils may be used. In the case of two or more pairs, not only the magnitude of the correction magnetic field but also the direction can be freely set, which is convenient.

  As a result, even in an apparatus employing an electron irradiation system equipped with an objective lens that leaks a magnetic field on the sample side, it is possible to irradiate the sample without separating the Ga isotope ion beam. Therefore, it is possible to achieve both the higher fine processing capability and higher resolution in STEM observation in one sample chamber.

1: electron gun, 2: focusing lens, 3: electron beam deflector, 4: objective lens, 5: electron beam irradiation system, 6: electron beam, 7: sample, 7a: thin film portion of sample, 8a: transmitted electron beam 8b: scattered electron beam, 9a: transmitted electron beam detection means, 9b: scattered electron beam detection means, 11: sample chamber housing, 12: ion gun, 13: focusing lens, 14: ion beam deflector, 15: objective Lens: 16: Ion beam irradiation system, 17: Ion beam, 18: Exhaust system, 19: Secondary electrons, 20: Secondary electron detection means, 21: Secondary electron entrance surface of secondary electron means, 22: Sample holder , 23: Sample holder axis, 24: Reflected electron, 25: Reflected electron detector, 26: Sample stage, 27: Correction magnetic field generator, 30: Intersection of irradiation axis of ion beam and electron beam, 40: Beam control means, 41: Image display means, 42: exhaust system control means, 50: linear axis drawn from the sample to the secondary electron detection surface (in the case of one sheet)

Claims (8)

An electron beam irradiation system for narrowing down the electrons emitted from the electron gun as an electron beam and irradiating the sample while scanning;
A transmission electron detector that detects a low-angle scattered electron beam transmitted through the sample;
A scattered electron detector for detecting a high-angle scattered electron beam transmitted through the sample;
Ion beam irradiation system that narrows the ions emitted from the ion gun as a focused ion beam, irradiates the sample while scanning, and processes the sample;
A secondary electron detector for detecting secondary electrons emitted from the sample by irradiation of at least one of the electron beam and the focused ion beam;
During processing of the sample by the focused ion beam, a bright field image or dark image of a scanning transmission electron microscope (STEM) of the sample using detection signals of the transmission electron detector, the scattered electron detector, and the secondary electron detector. Image display means for displaying at least one of a field image or a scanning electron microscope (SEM) image;
A sample holder for holding the sample when the sample is processed by the ion beam irradiation system;
The charged ion beam apparatus in which the irradiation axis of the focused ion beam and the irradiation axis of the electron beam intersect substantially at right angles, and the sample holder is disposed at the intersection position,
Control means for calculating a contrast between the transmitted electron detector signal and the scattered electron detector signal and stopping irradiation of the focused ion beam when the calculated contrast value reaches a predetermined threshold value. Charged particle beam device.   2. The charged particle beam apparatus according to claim 1, wherein the electron beam irradiation system has a magnetic field type objective lens, and a lens magnetic field of the objective lens leaks to a position of the sample holder. Beam device.   3. The charged particle beam apparatus according to claim 2, wherein the objective lens is an objective lens that forms a lens magnetic field with a magnetic yoke, and the magnetic yoke is disposed only on the incident side of the electron beam with respect to the sample holder. A charged particle beam device.   2. The charged particle beam apparatus according to claim 1, wherein the axis of the sample holder is on an axis that is perpendicular to the irradiation axis of the electron beam and the irradiation axis of the focused ion beam and passes through the intersection of both irradiation axes. Characteristic charged particle beam apparatus.   2. The charged particle beam apparatus according to claim 1, wherein the ion beam irradiation axis and the electron beam irradiation axis are orthogonal to a plane, and substantially symmetrical with respect to a plane including the ion beam irradiation axis. A charged particle beam apparatus, wherein a secondary electron entrance surface of a secondary electron detector is disposed.   2. The charged particle beam apparatus according to claim 1, wherein the image display means displays the sample scan image by the ion beam and the sample scan image by the electron beam by dividing the region into one display image frame. A charged particle beam device.   7. The charged particle beam apparatus according to claim 6, wherein the image display means displays the contrast of the sample scan image by the electron beam in time series as the sample is processed. The charged particle beam device according to claim 7.
Registration means for registering the contrast threshold;
Control means for controlling the irradiation of the ion beam,
The charged particle beam apparatus, wherein the control unit stops irradiation of the focused ion beam when a contrast of a sample scanning image by the electron beam reaches a threshold value registered by the registration unit.