JP5427569B2 - Charged particle beam scanning method and charged particle beam apparatus - Google Patents

Description

  The present invention relates to a charged particle beam scanning method and a charged particle beam apparatus, and more particularly to a scanning method and a charged particle beam apparatus capable of suppressing the influence of charging.

  When an electron beam is applied to the sample, secondary electrons are generated. In a scanning electron microscope (Scanning Electron Microscope: SEM), an observation image of a sample surface is obtained using a phenomenon in which the amount of generation varies depending on the shape of the sample. In image formation, scanning is performed for each line (horizontal line in the screen) in the horizontal direction in the screen, that is, in the raster direction. Furthermore, in order to improve the S / N of the image, not only one scan but also a plurality of images obtained by repeating scanning image acquisition in the same visual field usually several times to several tens times are added together. . An image obtained by these one scans is called one frame, and adding them together is called frame integration. Here, in the conventional scanning electron microscope, the scanning order of each line is exactly the same in both the first frame image and the second and subsequent frame images.

  According to the disclosure in Patent Document 1, it is explained that by devising the scanning order of the charged particle beam, the influence of charging that is received between the lines while the image for one frame is being captured is reduced. ing. Japanese Patent Application Laid-Open No. 2004-228561 describes a method of performing beam scanning in different directions for each frame. Further, Patent Document 3 describes that the scanning position is randomly changed.

JP 2007-059370 A (corresponding US Pat. No. 7,187,345) JP 2009-037804 A (corresponding US Patent US 2009/0032723) JP 05-151921 A (corresponding US Pat. No. 5,818,217)

  In the raster scan method, scanning is repeated in the same order between frames over and over, so that a charging phenomenon due to electron beam irradiation on the sample surface peculiar to the order is generated. The inventors have clarified that, depending on the sample, this charging phenomenon occurs with an excessive bias, which may affect the image quality of the scanned image by the electron beam.

  That is, when a line is being scanned, the charge remaining on the line that has already been scanned immediately before that affects the primary and secondary electron beams that are being scanned, thereby changing their trajectory. The phenomenon is happening. When this phenomenon occurs, when a certain line in the image is scanned, in the prior art, the line that was scanned immediately before is also the same, so that the influence of charging received from them is accumulated.

  According to the technique disclosed in Patent Document 1, accumulation of charge can be alleviated by optimizing the scanning order within one frame. However, the remaining charge exceeds the image capture time of one frame. There is no discussion about the effects of. Further, according to the technique disclosed in Patent Document 2, there is described a technique for reducing image drift caused by charges accumulated by electron beam scanning by scanning a beam in a different direction for each frame. Since the last scan line trajectory of the electron beam before the frame and the first scan line trajectory of the frame to be newly scanned are at the same position, depending on the relaxation time of charging caused by the electron beam irradiation one frame before, the new scan line May be affected.

  Furthermore, Patent Document 3 describes a technique for suppressing partial accumulation of charging by randomly scanning an electron beam. However, even if the same target sample is scanned because it is random, There is a possibility that the image and measurement result obtained each time the image is acquired.

  Hereinafter, a charged particle beam scanning method and a charged particle beam apparatus aiming at coexistence of suppression of the influence of charging caused by scanning of the previous frame and suppression of charging phenomenon inherent to scanning patterns caused by scanning in the same order Will be explained.

  As an aspect for achieving the above object, in a charged particle beam scanning method and apparatus for scanning a charged particle beam with a predetermined scanning pattern, a scanning start point in a vertical direction of a scanning line is set in a predetermined direction according to frame update. In addition, a method and an apparatus for changing by a predetermined amount are proposed.

  According to the above configuration, the scanning pattern can be changed for each frame while avoiding the final scanning line position where charging is considered to be most accumulated by beam scanning in the previous frame. It is possible to form a charged particle beam image without any bias.

Schematic of a scanning electron microscope. The figure which shows an example of the image memory of a scanning electron microscope. The figure explaining an example of a deflection pattern. The figure explaining an example of a deflection pattern. The figure explaining transition of the scanning start point when changing the scanning order for every frame. The figure explaining the other Example which moves a scanning start point for every flame | frame. The figure explaining further another Example which moves a scanning start point for every flame | frame. The figure explaining further another Example which moves a scanning start point for every flame | frame. The figure explaining the Example which uses together the scanning method which moves a scanning start point for every flame | frame, and the scanning method which can relieve the influence of charging within one frame.

  In the following, when capturing the second one-frame image after capturing the first one-frame image, a scanning line order different from the order of the scanning lines used when acquiring the first frame image is set. An example will be described. Further, when the third one-frame image is captured, a scanning line order different from the scanning line order used when the first frame and the second frame are acquired is set. A charged particle beam scanning method and apparatus characterized by scanning similarly thereafter will be described with reference to the drawings.

  According to the above configuration, the charging bias caused by the same scanning order does not accumulate during the frame integration between the first, second, and third frames. Therefore, the influence of the residual charging can be suppressed.

  According to the scanning order determination method as described above, it is possible to form an image with no brightness deviation in the screen.

  In this embodiment described below, in order to prevent an unnatural distribution in brightness and contrast in the sample image due to the influence of local surface charging, the first frame image, the second frame image, and thereafter A method of scanning each frame image in such a manner that the raster scanning order in the vertical direction in the screen is different from each other will be described.

  By doing in this way, a favorable image can be taken with the sample made from the following two types of materials. In other words, in the case of a material that remains charged for a long time required to capture an image for one frame on the already scanned line, the effect of the charge remaining on the line in the previous frame is different. Relax without accumulating between frames in order.

  Hereinafter, an outline of the SEM will be described with reference to the drawings. FIG. 1 is a schematic configuration diagram of an SEM. In this embodiment, an example of a scanning electron microscope that scans an electron beam to form a two-dimensional image of a sample will be described. However, the present invention is not limited to this. It is also possible to apply to a FIB (Focused Ion Beam) apparatus that scans a beam and forms a SIM (Scanning Ion Microscope) image.

  In the present embodiment, as shown in the figure, the electron beam source 1, the first convergent lens 3 and the second convergent lens 4 that converge the primary electron beam 2 emitted from the electron beam source 1, and the primary electrons. A deflector 5 that deflects the line 2 to scan on the surface of the sample 7, an objective lens 6 that focuses the primary electron beam 2 on the surface of the sample 7, and the primary electron beam 2 is the sample 7. A secondary electron detector 12 for detecting secondary electrons 16 generated after colliding with the surface, a first convergent lens power supply 8 and a second convergent lens power supply for driving the first convergent lens 3 and the second convergent lens 4 9, a deflection signal generator 11 that generates a deflection signal so that the primary electron beam 2 is scanned on the surface of the sample 7 by a predetermined method, and a deflector drive that drives the deflector 5 in response to the deflection signal. And secondary power detected by the secondary electron detector 12 An amplifier 13 that amplifies the signal, an image construction device 18 that generates an image from the amplified secondary electron signal, and the objective lens 6 are driven so as to focus the primary electron beam 2 at a predetermined position. An objective lens power supply 14 and a controller 15 that performs the above control are included.

  Here, when the primary electron beam 2 first scans on the sample 17 on the line 17a in the first frame acquisition, it then scans on the line 17b. Next, the line 17c is scanned.

  After the first frame acquisition, when the line 17a in the second frame acquisition is similarly scanned, the line 17b is next scanned. Next, the line 17c is scanned. However, the lines a, b, and c at this time are selected so as to be located on different lines from the first frame acquisition.

  After the second frame acquisition, when the line 17a is scanned in the third frame acquisition, the line 17b is scanned next. Next, the line 17c is scanned. However, each of the lines a, b, and c at this time is selected so as to be located on a different line from the first frame acquisition and the second frame acquisition.

  Next, detailed operations of the deflection signal generator 11 and the image construction device 18 will be described with reference to FIG. First, in the deflection signal generator 11, in accordance with the clock output from the write clock output circuit 21, the address generation circuit 22 outputs an address indicating the position where the sample 7 is irradiated with the electron beam. Based on the address, analog signals corresponding to the amount of deflection of the primary electron beam 2 from the D / A converter 23 are generated in each of the horizontal direction and the vertical direction. The deflector driver 10 drives the deflector 5 in accordance with the analog signal.

  One image construction device 18 operates as follows. That is, the secondary electron signal detected by the secondary electron detector 12 is amplified by the amplifier 13 and then converted into a digital signal by the A / D converter 24. This digital signal is stored in a memory group in the image memory 26 indicated by the input switch 25. The memory group selected here has a one-to-one relationship with the line indicated by the address generated by the address generation circuit 22. Here, the deflection position of the primary electron beam generated by the address generation circuit 22 is controlled according to the deflection pattern A shown in FIG. Therefore, the image data for each line sent to the image memory is not necessarily arranged in order from top to bottom in the vertical direction (perpendicular to the line) in the observation region on the sample. The input switch 25 also rearranges these image data according to the deflection pattern A so that these image data are arranged in order in the vertical direction from top to bottom in the image memory.

  When such a procedure is repeated and an image corresponding to a predetermined area on the sample 7 is taken, the image data stored in the image memory 26 is displayed according to the following procedure. That is, in accordance with the clock output from the read clock output circuit 27, the read address generation circuit 28 outputs an address indicating a drawing position on the display 33. Based on the address, an analog signal corresponding to the amount of deflection of the drawing electron beam generated in the display 33 from the D / A converter 29 is generated in each of the horizontal direction and the vertical direction. In accordance with the analog signal, the deflector 30 drives the deflector in the display 33.

  At this time, the image construction device 18 operates as follows. That is, with the image data already stored in the image memory 26, the image data for one line is read as a digital signal from the memory group in the image memory 26 indicated by the output switch 31. The memory group selected here has a one-to-one relationship with the line indicated by the address generated by the read address generation circuit 28. The read digital signal is converted into an analog signal by the D / A converter 32 and supplied into the display 33. In the display device 33, the luminance of the drawing electron beam generated from the cathode is changed in accordance with the analog signal, and the drawing electron beam is deflected by the deflector in the display device 33 described above. To display the image.

  Here, the drawing position on the display 33 generated by the read address generation circuit 28 is controlled according to the deflection pattern B shown in FIG. Therefore, the image data for each line output from the image memory is arranged in order from top to bottom in the vertical direction (perpendicular to the line) in the observation region on the sample. The output switch 31 sends these image data to the display device 33 in order according to the deflection pattern B. Through these processes, a scanning electron microscope image of the sample is displayed.

  Separately from this figure, the image data stored in the image memory 26 is sent to the controller 15 to achieve a predetermined purpose through appropriate image processing.

  Here, the deflection pattern A will be described with reference to FIG. In the scanning electron microscope of the present invention, scanning is performed at a constant cycle according to a sawtooth waveform in the horizontal X direction. This is called a raster scan. On the other hand, in the vertical Y direction, the scanning position is controlled in accordance with the raster scanning cycle. In the case of the deflection pattern A, the scanning line position “a” in the period a), the scanning line position “b” in the period b), the scanning line position “c” in the period c), the scanning line position “d” in the period d). , Cycle e), the line on the scanning line position “e” is scanned.

  The deflection pattern B will be described with reference to FIG. According to the deflection pattern B, the cycle a) → b) → c) → d) → e) proceeds in exactly the same manner as that used for the primary electron beam deflection in the conventional scanning electron microscope. Deflection is performed on the line or the line located at the position moved in steps at a certain distance of two lines.

  With the above configuration, it is possible to realize a scanning electron microscope capable of suppressing the bias of the charging phenomenon in the vertical direction with the raster scan generated by scanning the electron beam sample. In the case of this example, since the scanning is performed in a different order from the scanning order in the already scanned frames, an observation image that is a combination of the finally obtained frame images is obtained by charging the entire scanning region. It becomes possible to suppress the obstruction of the image and to effectively prevent the bias of the charging phenomenon.

  Next, a charged particle beam scanning order setting method and another scanning order setting method will be described with reference to the drawings different from the above-described embodiment. The charged particle beam scanning method described below is performed by scanning a charged particle beam in a two-dimensional direction in the X direction (scan line direction) and the Y direction (direction perpendicular to the scan line). The signals are integrated to form an image. Specifically, when scanning a plurality of frames, the scanning start point in the Y direction is changed according to the update of the frame, and the change is changed by a predetermined amount in a predetermined direction. FIG. 5 is a conceptual diagram of the charged particle beam scanning method described in this embodiment. In FIG. 5, for ease of explanation, a single field of view (FOV) is scanned five times in the X direction while being synchronized with scanning in the Y direction, and the scanning of the one frame is performed five times. The method of repeating is illustrated. The image signal for five frames becomes one image by image integration. Note that, in a charged particle beam apparatus such as a scanning electron microscope, an image is actually formed by hundreds of scanning lines.

  In the example of FIG. 5, the starting point of the scanning line in the Y direction is changed downward by an interval between scanning lines in accordance with the frame update. Furthermore, the position of the scanning line to be scanned in another order is also moved in accordance with the movement of the scanning start point so that the entire scanning in the FOV is possible.

  A general raster scan starts scanning from the scanning start point (scanning line position scanned at the top of the FOV) and scans at the scanning end point (scanning line position scanned at the bottom of the FOV). Exit. On the other hand, in the raster scan, the beam is scanned from the upper side to the lower side of the FOV. Therefore, the charge accumulated by the beam scanning is accumulated cumulatively as it goes below the FOV. However, the charging due to the beam scanning is not permanently accumulated, and gradually disappears. Therefore, at the end of one frame scanning, more charges are accumulated in the lower part of the FOV than in the upper part of the FOV. is there. In such a state, the position where beam scanning can be performed without receiving the charge generated by scanning the previous frame is the top of the FOV. In other words, the repetition of the raster scan can be regarded as a method that does not leave the influence of the charging of the previous frame in the scanning of a new frame when scanning a plurality of frames.

  On the other hand, recently, when scanning is continued with the same scanning pattern, a charging phenomenon depending on the scanning pattern has come to appear. This phenomenon is considered to change depending on the type of material of the sample used for, for example, a semiconductor device. That is, when acquiring the image signal of a plurality of frames, if the same raster scan is repeated a plurality of times, a charging phenomenon peculiar to the scan pattern may occur.

  As described above, the raster scan when acquiring a plurality of frames can suppress the accumulation of charge between frames, but causes a charge bias inherent to the scan pattern. The scanning method illustrated in FIG. 5 is for suppressing the bias of charging depending on the scanning pattern while following the effect of suppressing the charge accumulation of the raster scan. Specifically, the scan start position of the next frame is shifted with respect to the scan start position considered to be the place where the charge generated by the scan of the previous frame is the least, and a predetermined limit is provided for the shift amount. Thus, it is possible to eliminate the charge bias caused by the continuous scanning with the same scanning pattern while suppressing the accumulation of the charge between the frames.

  In the case of the method illustrated in FIG. 5, in the first frame (FIG. 5A), beam scanning is performed on five orbits (1) to (5) in order from the top. A white arrow indicates the first scanning position of each frame. At the end of scanning of the first frame, the position where the accumulation of charge is considered to be the least is the uppermost part in the FOV, but in the second frame, the uppermost part of the FOV is scanned by the first scanning line. If it is assumed that the raster scan is performed, the first frame and the second frame have the same scanning pattern. Therefore, in this example, as illustrated in FIG. 5B, the scanning is started by shifting the scanning start point by the interval between the scanning lines. Then, every time scanning of each frame is finished, as shown in FIGS. 5C, 5D, and 5E, the scanning start point is shifted by one scanning line interval, and scanning of all the frames is performed. Do.

  In the scanning method illustrated in FIG. 5, the scanning pattern is changed for each frame, but the scanning order does not change irregularly as in random scanning, and the scanning position is shifted with regularity in one direction. Therefore, it is possible to prevent a bias in charge while maintaining high image reproducibility.

  FIG. 5 illustrates an example in which the scanning start point is shifted by one scanning line interval for each frame. However, the present invention is not limited to this. For example, the scanning start point is moved in units of a plurality of scanning line intervals. You may make it do.

  In the method illustrated in FIG. 5, the example in which the scanning end point is also moved in accordance with the movement of the scanning start point has been described. However, the present invention is not limited to this. For example, as illustrated in FIG. The scanning end point position may be unified regardless of the change. In the case of FIG. 6, the lowermost portion of the FOV is the scanning end point common to each frame. In this case, as the frame is updated, the acquired image becomes smaller (the number of scanning lines decreases), but the measurement or inspection target region 607 (ROI: Region Of Interest) is acquired without omission in each frame. If so, pattern measurement and inspection can be performed with high accuracy.

  FIG. 6 exemplifies a method of integrating the image signals of the first frame 601, the second frame 602, the third frame 603, the fourth frame 604, and the fifth frame 605 to form an integrated image 606. The first frame 601 to the fifth frame 605 have scanning lines 609, 610, 611, 612, and 613 as scanning start points, respectively. Further, although the scanning areas of the respective frames are different in size, signals based on the respective scanning lines are integrated in an area below the scanning line 613 that is an effective visual field area by a control device described later. According to such a method, it is possible to appropriately form an image of a region to be measured or inspected while suppressing the bias of charging depending on the scanning pattern.

  Further, FIG. 7 is a diagram for explaining an example in which the position of the FOV is changed in accordance with the movement of the scanning start point. In this example, the measurement target is a line pattern 701 that is long in the vertical direction (direction perpendicular to the scanning line direction). In addition, the scanning start point is moved at an interval between three scanning lines (interval between two scanning lines: Δy). In this example, an example in which an image is formed with three frames of (a) the first frame, (b) the second frame, and (c) the third frame is described. However, the present invention is not limited to this. The number of frames can be set arbitrarily.

  In this example, since the beam scanning is performed so as to trace the same scanning line trajectory within the effective field of view while shifting the scanning start point, it is possible to perform image integration while maintaining high sharpness. In the case of this example, since the scanning order at each scanning position changes with the shift of the scanning start point, signals obtained at each scanning position (Secondary Electron (SE) and backscattered electrons (Backscattered) When storing Electron: BSE)) in a storage medium such as a frame memory, the storage parts are rearranged in units of frames.

  Further, as illustrated in FIG. 8, the same scanning may be repeated while shifting the position of the FOV in the Y direction. Also in this case, the scanning start point is moved in units of a predetermined interval between scanning lines. FIG. 8 is a diagram illustrating an example in which a field of view for acquiring an image of a region including four hole patterns is set for a sample in which a plurality of hole patterns are arranged. In the example of FIG. 8, in order to acquire information on the dimension Rx in the Y direction, a scanning range larger than the dimension Rx is set in the Y direction. In this case, the scanning line is scanned by “the number of scanning lines necessary for scanning the dimension Rx + (the number of necessary frames fn × the interval Sd between scanning lines Sd) −1”, and the scanning of the one frame is performed. Repeat as many times as necessary. Specifically, when the required number of frames is 4 and the scanning start point movement amount between the frames is an interval between one scanning line, the number of scanning lines required to scan the dimension Rx + 3 scanning lines. By performing scanning for four frames, it is possible to change the scanning pattern for each frame while obtaining information necessary to form an image of the dimension Rx.

  In the scanning method as described above, the uppermost scanning line of the first frame is the fourth scanning line scanned in the frame, and the uppermost scanning line of the second frame is the third scanning line in the frame. A scanned scan line is obtained. That is, the scanning order is changed for each frame. Thus, by setting the scanning range, it is possible to acquire an image with an effective field of view while changing the scanning pattern for each frame.

  Next, an example in which a method of changing the scanning order of scanning lines for each frame is applied to a scanning method other than raster scanning will be described. FIG. 9 is a diagram for explaining an example in which the method of changing the beam scanning order in units of frames described above and the floor plan scanning disclosed in Patent Document 1 are used together. In Patent Document 1, by setting a beam scanning pattern so that the next scanning line is scanned at the center between the already scanned scanning lines, the influence of charging generated by the already scanned scanning lines is disclosed. Suppressing scanning methods are described. Since the position where the beam is newly scanned is located at the center of the already scanned scanning line trajectory, the new beam is scanned at a position where the influence of charging is balanced, which is brought about by the already scanned scanning line. It is possible to suppress the influence of charged electric charges. For example, in the first area of FIG. 9, the highest scanning position is scanned first, and then the lowest scanning position of the first area is scanned. Thereafter, in the first region, a scanning line is set so that a new scanning line is scanned at the center of the already scanned scanning line trajectory interval, and this is repeated. When the scanning order is not changed for each frame, the scanning to the scanning positions (1) to (17) is performed in numerical order from the first area to the nth area.

  FIG. 9 is a diagram for explaining an example in which a technique for changing the scanning order for each frame is applied to such a scanning method. Specifically, one FOV is divided into a plurality of regions (first region, second region...), And the beam is set to be scanned in a different scanning order for each frame. More specifically, each time the frame is updated, the scan start point is set so that the scan start point is shifted by one region (by an interval between 16 scan lines). According to such a scanning order, it is possible to realize both prevention of charging bias by changing the scanning start point for each frame and mitigation of charging influence by the existing scanning line within one frame.

DESCRIPTION OF SYMBOLS 1 Electron beam source 2 Primary electron beam 3 First convergent lens 4 Second convergent lens 5 Deflector 6 Objective lens 7 Sample 8 First convergent lens power supply 9 Second convergent lens power supply 10 Deflector driver 11 Deflection signal generator 12 Secondary electron detector 13 Amplifier 14 Objective lens power supply 15 Controller 16 Secondary electrons 17a, 17b, 17c Line 18 Image construction device 21 Write clock output circuit 22 Address generation circuits 23, 29, 32 D / A converter 24 A / D converter 25 Input switch 26 Image memory 27 Read clock output circuit 28 Read address generation circuit 30 Deflection amplifier 31 Output switch 33 Display

Claims (5)

In a charged particle beam scanning method of acquiring image signals for a plurality of frames by scanning a charged particle beam with respect to a two-dimensional region on a sample and integrating the image signals of the plurality of frames,
In response to said update accumulated the frame, the charged particle beam scanning method, wherein the benzalkonium moving the scanning start point interval basis between predetermined scan lines. In claim 1,
With the movement of the scanning start point, the charged particle beam scanning method, characterized in that the scanning order is changed for each scanning position before notated in frame. A charged particle beam comprising: a deflector that scans a charged particle beam on a sample; a control device that controls the deflector; and a storage medium that stores an image signal obtained by scanning the charged particle beam with respect to the sample in units of frames. In the device
The control device controls the deflector so as to move the scanning start point in units of a predetermined interval between scanning lines in accordance with the update of the accumulated frame, and the image signal stored in the storage medium for each frame Are charged to form a composite image. In claim 3,
With the movement of the scanning start point, before the charged particle beam apparatus characterized by scanning order is changed for each scanning position of notated in frame. In claim 3,
The charged particle beam apparatus, wherein the control device controls the deflector so that a scanning start point moves in a direction perpendicular to the scanning line direction.

Images