JP2009103470A - Method for inspecting scanning accuracy of scanning irradiation device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for easily inspecting the scanning accuracy of a scanning irradiation device by using a moire method for enlarging a slight difference in interval and showing it, especially an electron beam moire method excellent in measurement in a minute region. <P>SOLUTION: An image is obtained by making a board and a material for grids different in permeability, or reflection characteristics, or electron/ion generating characteristics to a particle beam or an energy beam, and emitting the particle beam or energy beam in parallel or with an acute angle to a reference board as an inspection sample acquired by forming grids on the board, in such a way as to perform scanning. A grid to be a reference is manufactured so as to be proportionate to the capability of a device to calibrate the grid so as to perform inspection of a wide range of magnification, and by using the grid, moire fringes are generated. By the shape of these moire fringes, ununiformity and inaccuracy of a scanning width are inspected and calibrated. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a scanning accuracy verification method for a scanning irradiation apparatus.

Conventionally, the following methods have been known as methods for verifying the scanning accuracy of irradiation rays.
In order to verify the magnification of the scanning electron microscope, a “magnification standard sample”, which is a standard for magnification, having a certain size or pitch is used. There are meshes, latex spheres, etc. Recently, patterns with a pitch of several hundred nm to several tens of μm using a lithography technique have been used. In order to verify, the standard sample was observed at a certain magnification, the length of the reference object was measured, the magnification was corrected, and this observation was used to investigate the accuracy of the scanning position and width based on image distortion and aspect ratio differences. .
In Japanese Patent Publication No. 2535787, information on the height of the probe is changed to information on luminance using atomic molecules existing in the natural world, and moire fringes are observed, which is used for calibration of linearity of the scanning probe microscope. is described. Since a natural object is used, only a lattice of a region existing in the natural world can be used, so that a region to be calibrated is limited. Also, with this method, it is difficult to detect when the distribution of scanning line intervals changes gently.
The method of obtaining the amount of deformation using moiré fringes is called the moire method, and the optical moire method is a well-known method for a long time, where two lattices (model lattice and master lattice) are overlapped and overlapped. And the amount of deformation is determined from light and dark stripes (moire stripes) formed by portions that do not overlap. Furthermore, the inventors have also developed an electron beam moire method (18775579). In the electron beam moire method, a model lattice is prepared using a material having a different amount of secondary electrons from that of the sample, and an electron beam irradiated in a lattice shape is used instead of the master lattice, and the electron beam is placed on the model lattice. In this method, moire fringes are generated based on the difference in the amount of secondary electrons and reflected electrons generated upon irradiation. Compared to the conventional optical moire method, it is possible to measure the amount of deformation in a minute region.
Japanese Patent No. 2535787 Japanese Patent No. 1875579 http://www.jeol.co.jp/cgi-bin/searchDB.pl?query=Φ&searchWhat=semTerms

  As described above, the method for verifying the scanning accuracy of the scanning irradiation apparatus uses a sample as a reference for magnification, observes the standard sample using the apparatus, measures the length of the reference object, corrects the magnification, Since the accuracy of the scanning position and width was investigated by the difference in distortion and aspect ratio, it took time to repeat the adjustment and observation, and there was a possibility that it would be overlooked in the case of minor deviations.

Therefore, we developed a method for easily verifying the scanning accuracy of a scanning irradiation device by using the moire method that expresses a slight difference in spacing, especially the electron moire method that excels at measurement in a fine area. Tried. Here, in order to be able to test the magnification in a wide range, instead of a naturally existing grid as shown in Japanese Patent Publication No. 2535787, a grid that accurately matches the ability of the device to calibrate the reference grid (with high accuracy) Moire fringes are generated using this, and the non-uniformity or inaccuracy of the scanning width is verified and calibrated from the shape of the moire fringes.
That is, according to the present invention, as a test sample, the grid is formed on the substrate such that the material forming the grid and the substrate have different transmission or reflection characteristics with respect to particle beams or energy beams, or generation characteristics of electrons and ions. An image is obtained by irradiating the particle beam or energy beam in a scanning manner in parallel or with an acute angle to the reference plate.

  By this method, it is possible to easily perform a scanning accuracy test of an apparatus that scans and irradiates particle beams and energy beams, and it is possible to detect a minute error that has been overlooked so far. Further, since the shift of the parallel grating group can be enlarged and displayed as shown in FIG. 13, it can be used for alignment of the stage in the apparatus for scanning the beam. If the moiré fringes match, it indicates that the parallel straight line group is perfectly aligned and can be easily aligned.

In the present invention, since the master grid serving as a reference is artificially and accurately manufactured, the scanning width can be verified in a wide range (from micron order to millimeter order).
The grid is manufactured using photolithography, electron beam lithography, X-ray lithography, a laser processing machine having a high-precision XY stage, or the like. A grid suitable for the capability of the apparatus to be calibrated is produced accurately (with high accuracy), and moiré fringes are generated using this grid to calibrate the nonuniformity and inaccuracy of the scanning width. The grid interval is preferably the same as or slightly different from the scan interval that the apparatus should originally have. In the case of a grid that is substantially the same as the scanning interval that the apparatus should originally have, moiré fringes are observed from 0 to 1 on the screen, and in the case of slightly different intervals, it is easy to judge whether several to a dozen or so are observed. Different moiré fringe intervals mean different particle beam scanning intervals.
Fig. 4 shows the principle of moiré fringe generation. Moire fringes are fringes formed by overlapping figures with two similar patterns.
As shown in the left diagram of FIG. 4, moire fringes generated when the grid intervals are different are referred to as mismatched moire fringes, and moire fringes that appear when the two grids shown in the right diagram have a certain angle are referred to as misaligned moire fringes.

When measuring strain or the like, a straight line group or a perpendicular grid parallel at equal intervals is used. When the interval of the master grid of the reference grid is a and the interval of the model grid (deformable grid) is a ′, the change rate (strain) ε of the model grid with respect to the master grid is expressed by Equation 1.
(Formula 1)


The Moire method is a technique that uses several straight lines to obtain strain as in Vernier calipers, and is a technique that improves the average strain measurement accuracy at the expense of spatial resolution. The scanning width and direction of the apparatus that irradiates lines (photons, electrons, ion beams) in parallel lines can be obtained from the change (interval and angle) of moire fringes. This phenomenon is used to test the scan width and angle of an apparatus that regularly scans all particle beams and energy beams.

Since the present invention is a test method for the scanning width of microscopes that irradiates (scans) particle beams and energy beams at equal intervals on a test piece, a grid in which a master grid is accurately prepared is used, and scanning light and electron beams are used. Model grid. When a laser or light is used, a grid that reflects light differently from the substrate is used as the master grid (Fig. 5), and when the electron beam is irradiated, the amount of secondary electrons generated is different from that of the substrate. Use the grid (Fig. 6).
Like a scanning acoustic microscope, high-frequency ultrasonic waves (100 MHz to 3 GHz) are focused using an acoustic lens, etc., and the sample placed on the focal plane is mechanically scanned, and the partial ultrasonic reflectance or transmission of the sample is obtained. In the case of a device that detects a change in the rate and displays it as an image on a CRT, a master grid manufactured using materials having different elastic moduli and different densities is used.

  When a grid (master grid) having the same scanning width as that of the correct scanning width is prepared in advance, and when this grid is set parallel to the scanning width of the particle beam and observed, the moiré fringes cannot be observed at all. The scan width is the correct magnification. Consciously irradiating particle beams and energy beams (photons, electrons, ion beams) in parallel lines at intervals different from the grid (master grid), and when the moire fringes are equally spaced, they are scanned at approximately equal intervals. However, if there is a difference in the moire fringe interval, it indicates that the scanning width is not equal.

  When it is possible to leave the sample after irradiation with an apparatus having no display unit such as a laser processing machine, moiré fringes generated between the grid after irradiation with a sample having a grid may be used. Also, the angle between the grid (master grid) and the particle beam or energy beam group, but the distance between the moire fringes becomes wider when scanned parallel to each other or at an acute angle. Easy to detect. When the angle exceeds 30 degrees, the moire fringes hardly change, so 30 degrees or less is preferable.

FIG. 1 shows an orthogonal lattice with a spacing of 3 μm fabricated using a femtosecond laser irradiation apparatus.
The substrate is a laminate of 304 stainless steel and carbon steel. The grating was fabricated by placing the substrate on a 3-axis operation stage of a femtosecond laser irradiation apparatus and moving the stage in parallel at intervals of 3 μm while irradiating with the laser. The laser light source is IFRIT manufactured by Cyber Laser, and its wavelength is 780 nm. The pulse width for laser irradiation is 170 femtoseconds. Since the portion irradiated with this laser is heated to a high temperature, the substrate surface is ablated and fine irregularities are generated. Due to this unevenness, light is diffusely reflected and observed dark. This figure shows the results of observation with a scanning laser microscope (scanning laser microscope 1LM15W manufactured by Lasertec Corporation) using a 100 × objective lens.
FIG. 2 shows the results of moiré fringe observation using a 20 × objective lens with the microscope. The laser of this microscope (helium neon laser, wavelength 633 nm) is weak in output and can be used only for surface observation. In FIG. 2, when the interval of the irradiation laser is calculated from the magnification, the observation area, and the number of scanning lines, the irradiation is performed in parallel at an interval of about 2.2 μm.

  The moire fringes in FIG. 2 show straight lines, but the interval between the fringes is narrower at the top and slightly smaller at the center. If you go further down, it becomes narrower again. This indicates that the scanning interval, which should be at regular intervals, is wider in the central part than in the upper and lower parts, and narrower in the upper and lower parts than in the central part. This is because the galvanometer mirror for changing the laser irradiation angle is not operating normally.

  For comparison, a micromarker for an optical microscope having straight lines spaced at 10 μm was observed with a 20 × objective lens using the same scanning laser microscope (FIG. 3), but no significant change in the distance was observed. In this way, using the change in moire fringes makes it possible to detect the difference in distance much more sensitively than directly observing the marker.

  As shown in Table 1, other methods for creating a lattice on the substrate include electron beam lithography and photolithography, a method of creating by vapor deposition and etching, a method of creating grooves by ion polishing, a scanning probe microscope and an XY There are a method of making a sample by scratching it using a stage, and a method of making a metal by vapor deposition at a place irradiated with an ion beam by ion-assisted gas deposition.

The substrate is a laminate of 304 stainless steel and carbon steel. The grating was prepared by placing the substrate on a 3-axis operation stage of a femtosecond laser irradiation apparatus and moving the stage parallel to the x and y directions at 10 μm intervals while irradiating with the laser. The laser light source is IFRIT manufactured by Cyber Laser, and its wavelength is 780 nm. The pulse width for laser irradiation is 170 femtoseconds. The portion irradiated with this laser is heated to a high temperature, the substrate surface is ablated, and fine irregularities are generated. Due to this unevenness, light is diffusely reflected and observed dark. The produced orthogonal lattice is shown in FIG. FIG. 8 shows the results of observing moire fringes using a scanning laser microscope (scanning laser microscope 1LM15W manufactured by Lasertec Co., Ltd.) on this lattice, using a 5 × objective lens, and further adjusting the scanning line interval. When the magnification, observation area, and number of scanning lines are calculated, irradiation is performed in parallel at intervals of about 9 μm. Moire fringes show a straight line, but the spacing between the stripes is narrower at the top and larger at the bottom. This indicates that the scanning interval, which is originally not equal, is narrower on the photograph and wider on the bottom. This phenomenon is because the galvanometer mirror for changing the laser irradiation angle is not operating normally.
For comparison, a micromarker for an optical microscope in which straight lines with intervals of 10 μm were drawn was observed with a 5 × objective lens using the same scanning laser microscope (FIG. 9), but no significant interval change was observed. In this way, using the change in the moire fringes makes it possible to detect the difference in the interval between the laser beams scanned much more sensitively than when directly observing the marker.

  FIG. 10 shows the results of observation with a scanning electron microscope of 20 μm-interval parallel gratings produced using a femtosecond laser irradiation apparatus. The substrate was a laminate of 304 stainless steel and carbon steel. The grating was prepared by placing the substrate on a 3-axis operation stage of a femtosecond laser irradiation apparatus and moving the stage in parallel at 20 μm intervals while irradiating the laser. The laser light source is IFRIT manufactured by Cyber Laser, and its wavelength is 780 nm. The pulse width for laser irradiation is 170 femtoseconds. The portion irradiated with this laser is heated to a high temperature, the substrate surface is ablated, and fine irregularities are generated. Due to the unevenness, the amount of secondary electrons generated is increased and brightly observed. FIG. 11 shows an electron beam moire fringe observed by using a scanning electron microscope for the lattice and irradiating the electron beam parallel to the lattice at intervals of about 22 μm. The electron beam moire fringe spacing is narrow near the center and wide at both ends and up and down. Moire stripes are also curved. This indicates that the electron beam scanning interval, which should be at regular intervals, is narrow near the upper left and right ends and wide near the center. This is because in a scanning electron microscope in which the electron beam is controlled by an angle, the image is distorted because the angle and the scanning width do not match at low magnification. For comparison, FIG. 12 shows the results of observation at this magnification using a scanning electron microscope calibration pattern. Although concentric squares, which are originally squares, can be observed in the upper and lower portions of the photograph, the lines are slightly curved, but no significant change was observed as with moire fringes. In this way, using the change in moire fringes makes it possible to detect the difference in distance much more sensitively than directly observing the marker.

  This method is a scanning accuracy verification method for a scanning irradiation apparatus that irradiates a target with irradiation beams such as particle beams or energy beams in a scanning manner by using a lattice produced on a substrate accurately. A scanning irradiation apparatus that irradiates a target with irradiation beams such as particle beams or energy beams of this type is a microscopic device such as a scanning electron microscope, a scanning laser microscope, or a scanning scanning acoustic microscope in the embodiment. Not only can the image accuracy of the observation device in the area be verified, but also improve the accuracy of the drawing of the focused ion beam and laser processing machine for precision processing and the electron beam drawing device for making photomasks for making electronic components. Can contribute to the development of this field. Further, as shown in FIG. 13, since the shift of the parallel grating group can be enlarged and displayed, it can also be used for alignment of the stage in the apparatus for scanning the beam. Draw a parallel line at the end of the stage and draw a parallel line on the other side, draw parallel lines of the same width, irradiate particle lines and energy rays in parallel lines to generate moire fringes, and parallel if the moire fringes match Since it shows that the straight line group fits perfectly, it is possible to easily align. By using this method, the irradiation accuracy of particle beam or energy beam is improved and the positioning becomes easy, so it seems to contribute greatly to improving the yield in the field of electronic component manufacturing and precision processing. .

In addition, the general meaning of each term is described below as an aid for understanding the contents of the present invention.
Galvano mirror A unit for moving the optical axis to scan the laser at high speed. In general, two mirrors are moved by a motor, and the x and y directions on the sample are scanned. The operation is electrically controlled using a computer.

Scanning electron microscope A device that scans a target sample with an electron beam, detects reflected electrons and secondary electrons with a detector, displays the intensity as an image on a cathode ray tube, and obtains an enlarged image of the object. The reflection method (SEM) is widely used for observing surface shapes and tissues. Resolution is about 2-3mm. A method for observing a transmission electron image is called a scanning transmission electron microscope (STEM).

Scanning probe microscope A piezoelectric body that generates a very small deformation by applying a voltage to form a stage that can move in the X, Y, and Z directions, and scans a very small probe to capture data and image it. It is called SPM in the microscope. At present, a great many types have been devised. For example, among those already on the market, in addition to STM (scanning tunneling microscope) and AFM (atomic force microscope), LMFFM (Lateral Force Modulation Friction Force Microscope) that measures surface friction force, VE- that measures surface viscoelasticity There are AFM (Visco-elasticity Atomic Force Microscope), KFM (Kelvin Force Microscope) that measures surface potential distribution, and the like. In addition, as a major feature of SPM, it can be used in a very wide environment from vacuum to aqueous solution, and various processing as well as observation can be mentioned. As described above, SPM is very versatile.

Scanning acoustic microscope (ultrasonic microscope)
Focusing high-frequency ultrasonic waves (100MHz to 3GHz) using an acoustic lens, etc., mechanically scanning the sample placed on the focal plane, detecting changes in the partial ultrasonic reflectance or transmittance of the sample, A device that displays images on a CRT. Used for observing microstructures and minute defects of metals and ceramics. = SAM
Scanning laser microscope A microscope that scans laser light in parallel lines and displays the brightness of reflected light on a cathode ray tube for observation.

Magnification standard sample To confirm the magnification and performance of a scanning electron microscope, a magnification standard sample (a sample with a fixed size or pitch, which serves as a reference for magnification. To increase magnification accuracy, There are meshes, latex spheres, etc., but recently, patterns with a pitch of several hundreds of nanometers using lithography technology have been used to satisfy ISO traceability. Some use.)

FIG. 3 is a photograph showing the results of observation with a scanning laser microscope using a 100-times objective lens for the grid used in Example 1. FIG. 2 is a photograph showing an observation result of moire fringes by a scanning laser microscope using an objective lens 20 times the grid of FIG. The photograph which shows the observation result by the scanning laser microscope which used the micron marker for optical microscopes (minimum space | interval 10 micrometers) using a 20 times objective lens. The schematic diagram which shows the generation principle of a moire fringe. Left figure: Moire fringes due to different lattice spacing (mismatch). Right figure: Moire fringes due to non-parallel lattices (misalignment). The schematic diagram which shows the effect | action of the grid for microscopes which scans a laser or light. The schematic diagram which shows the effect | action of the grid for microscopes which scans an electron beam or ion. 6 is a photograph showing the results of observation with a scanning laser microscope using a 20 × objective lens for the grid used in Example 2. FIG. The photograph which shows the observation result of a moire fringe with the scanning laser microscope which used the objective lens of 5 times for the grid of FIG. The photograph which shows the micron marker observation result for optical microscopes. 6 is a scanning electron micrograph of a parallel grid with 20 μm intervals used in Example 3. FIG. Photograph showing electron moire fringes observed by irradiating the grid of FIG. 10 with an electron beam at 22 μm intervals using a scanning electron microscope 11 is a photograph showing a magnification calibration pattern for a scanning electron microscope observed at the same magnification as in FIG. The conceptual diagram of the moire fringe production | generation by the shift | offset | difference of a parallel lattice group. A slight shift can be enlarged and displayed as a large shift of moire fringes. The coincidence of moiré fringes means that two parallel straight line groups are exactly coincident.

Claims (1)

  A scanning accuracy verification method for a scanning irradiation apparatus that irradiates a target with irradiation beams such as particle beams or energy beams, and the grid and the substrate as the verification sample are transparent or reflective to particle beams or energy beams. Alternatively, the particle beam or the energy beam is irradiated in a lattice shape in parallel or at an acute angle with respect to the reference plate on which the grid is formed on the substrate so that the generation characteristics of electrons and ions are different. Characterized by