JP3802525B2 - Charged particle microscope - Google Patents

Description

  The present invention relates to a charged particle microscope used for observing a fine pattern of a semiconductor element, for example.

  In semiconductor development, observation of fine patterns with a scanning electron microscope (SEM) and fine processing techniques with a focused ion beam (FIB) are essential. In the microfabrication technology by FIB, in order to specify a processing target, prior observation by a charged particle microscope is necessary, and the importance of these microscopes is increasing more and more, but as the resolution becomes lower than 1 nm. There are many problems to be solved.

  FIG. 12 shows a schematic configuration of a conventional scanning electron microscope system. A focused electron beam is emitted from the electron optical column 1 around a point 13 (referred to as point A) on the wafer 3. The wafer is fixed on the stage 9. The stage 9 is placed on the horizontal movement mechanism 4 and the vertical minute movement mechanism 5, and by these mechanisms, movement in the horizontal direction (xy direction) and minute movement (Δz) in the vertical direction are possible. .

  These mechanisms are mounted on the first support body 6, and the tilt angle of the stage 12 around the rotation shaft 11 can be adjusted by the tilt mechanism 7 of the first support body 6. The first support 6 is held by the second support 8 via the tilt mechanism 7, and the position of the second support 8 in the vertical direction (z direction) is roughly adjusted by the vertical movement mechanism 19. Is done. Reference numeral 10 denotes a drive device for the vertical movement mechanism 19. The horizontal movement mechanism 4, the vertical minute movement mechanism 5, the tilt mechanism 7, and the driving device 10 are controlled by the stage controller 12. Reference numeral 2 denotes a housing of the entire system.

  Now, in many cases, it may be desirable to observe the sample tilted at, for example, 60 degrees. FIG. 13 shows the movement of the point A at this time. FIG. 13 shows the stage viewed in the direction of the central axis 11 (referred to as axis B) of the stage tilting mechanism 7.

  When the point A is on the axis B (FIG. 13A), the position of the point A does not change even if the surface 3a of the stage 3 is inclined by θ degrees to the position of 3b. However, when the height of the point A is higher than the axis B (11) (FIG. 13 (b)) or lower (FIG. 13 (c)), the point A changes from 13a to 13b in both the horizontal and vertical directions. Moving.

  Assuming that the inclination angle is θ, and the difference in height between the point A (13a) and the axis B (11) is h, the movement amount of the point A during the inclination is Δx = hsin (θ) in the horizontal direction and Δz = in the vertical direction. (1-cos (θ)) h. In many cases, the unevenness of the wafer 3 is about 100 μm. Therefore, when θ is 60 degrees, the movement amount of the point A is about 86 μm at the maximum in the horizontal direction and about 50 μm in the vertical direction.

  The amount of movement in the horizontal direction is much larger than the size of the electron beam observation region, for example, about 2 μm, and the amount of movement in the vertical direction is much larger than the focal depth of about 2 μm. Therefore, when the stage is tilted, it is necessary to move the point A to the center of the field of view by moving it horizontally by the horizontal movement mechanism and to adjust the vertical direction by adjusting the focus of the electron optical system. .

  Alternatively, the height of the stage 9 can be adjusted by the vertical minute moving mechanism 5 until the movement of the point A is minimized by inclining at a small angle. However, any of the methods has caused a significant decrease in the efficiency of observation work.

  The above is a problem of alignment in tilt observation, but image disturbance caused by fluctuation of the electron beam magnetic field or vibration is becoming a problem as the resolution of the microscope becomes less than 1 nm. Next, this problem will be described.

  FIG. 14 shows a configuration of a typical high-resolution scanning electron microscope. A field emission electron gun 25 generates an electron beam 28 having a very high luminance. Details of the periphery of the electron gun 25 are omitted. The electron beam 28 is further reduced by the condenser lens 26 and the objective lens 27 and irradiated on the surface of the sample 37. At this time, the secondary electrons 38 emitted from the sample surface are detected by the detector 33. Reference numeral 36 denotes an aperture.

  The electron beam 28 is scanned two-dimensionally on the sample 37 by the deflector 29. In synchronization with this, the signal of the detector 33 is displayed on the image display device 32, whereby information on the sample surface can be obtained. 31 is a sawtooth wave voltage generator used for scanning the deflection electrode 29 and the display device 32, and 30 is a power supply for the deflector.

  By the way, in the place where the scanning electron microscope is placed, there is usually an oscillating magnetic field outside. The floor is vibrating in some way. Alternatively, the atmosphere in which the device is located is also vibrating. Therefore, the position of the electron beam on the sample may be shifted from a predetermined position due to the influence of the oscillating magnetic field or vibration. Therefore, the image on the image display device 32 is different from the original structure on the sample 37.

  With respect to this problem, in Patent Document 1, an external magnetic field detection means (corresponding to 39 in FIG. 23) is provided, and the deflection signal of the electron beam is corrected based on the signal, thereby vibrating the external magnetic field. Regardless of this, a method has been proposed in which an electron beam is always applied to a predetermined position.

  On the other hand, Patent Document 2 proposes a method of obtaining an image with less distortion by inputting a correction signal in synchronization with a CRT deflection circuit of an image display device. However, this method has the following problems.

  It takes some time to detect the disturbance and generate the correction signal. Therefore, the correction of the beam position cannot be completely synchronized with the disturbance. When the disturbance changes quickly, the effect becomes significant, and the image signal may be disturbed by using a correction means having a time delay. Further, when the image display is simply corrected, a situation occurs in which the irradiation charge density of the beam varies depending on the location, and an apparent shading may occur.

  By the way, as an objective lens, it is known that a so-called in-lens or semi-in-lens system in which a magnetic field is also applied to the sample surface has a small amount of collection. The basic arrangement of the electron gun, the condenser, and the objective lens is the same as that in FIG. 14, but a deflector is incorporated in the objective lens.

  FIG. 15 shows an example of the structure of a semi-in-lens type objective lens. The opening part of the pole piece 52 faces downward. The left side of the figure shows an outline of the magnetic field strength distribution along the central axis of the lens. Since a magnetic field is applied to the surface of the sample 46, the trajectory of the secondary electrons 51 is limited to the vicinity of the center of the lens. Therefore, if the detector is attached outside the pole piece of the lens, sufficient detection efficiency cannot be obtained. Therefore, it is provided inside or upstream of the pole piece 52.

  The detector 58 is usually composed of a combination of a bias grid 54 for capturing electrons, a scintillator 55 and a photomultiplier tube 56. A shielding tube 57 is provided so as to surround the electron beam 42 so that the influence of the bias does not reach the electron beam. In this system, the electromagnetic deflector 44 is provided downstream of the detector 58 in order to increase the deflection region of the electron beam. However, in this case, the following problem occurs.

  FIG. 16 is a diagram for explaining this problem, and conceptually shows the trajectories of the electron beam 42 (primary electrons) and the secondary electrons 51 in a state where the deflector is working. The secondary electrons move while swirling, but for the sake of simplicity, the trajectory of the secondary electrons indicates the trajectory of the center of rotation. It is assumed that the magnetic field in the vicinity of the central axis is downward in the direction perpendicular to the paper surface (indicated by symbol 59).

In this case, the secondary electrons are deflected in the opposite direction to the primary electrons. Moreover, the energy of primary electrons is about 2 keV, for example, and the energy of secondary electrons is usually several tens eV at most, so the deflection angle itself is large. Therefore, when trying to widen the deflection region of the primary electrons, the secondary electrons are greatly deflected and lost along the way, and do not reach the detector. Therefore, when it is desired to increase the detection efficiency of secondary electrons, there has been a problem that the deflection region of primary electrons cannot be made very wide (cannot be scanned widely).
Japanese Patent Publication No.58-22854 JP-A-5-82068

  As described above, when the stage is tilted with a charged particle microscope, the region to be irradiated with the beam often deviates greatly in the horizontal and vertical directions, which significantly reduces the operability of the apparatus.

In charged particle microscopes, particularly scanning electron microscopes, it has been difficult to correct image disturbance due to fast-changing disturbances.
Further, in a scanning electron microscope using an in-lens type objective lens, when a deflector is provided downstream of the detector, it is difficult to widen the deflection region of the electron beam.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a charged particle microscope which has good operability and can provide a clear image over a wide range.

  More specifically, the first object is to provide a charged particle microscope in which the region to be irradiated with the beam does not move even when the sample stage is tilted. The third object is to provide a charged particle microscope capable of correcting disturbance, and thirdly, in the case where an in-lens type objective lens is used, even when a deflector is provided downstream of the secondary electron detector, the charging region can be widened. It is to provide a particle microscope.

  In order to solve the above-mentioned problems, in the charged particle microscope of the present invention, in the charged particle microscope which displays an image by detecting a secondary electron signal obtained by scanning an electron beam on a sample with an objective deflector using a detector. The detector is disposed upstream of the electron beam from the objective deflector, and the objective deflector includes an electromagnetic deflector and an electrostatic deflector that operates in synchronization with the electromagnetic deflector. The electric field of the electric deflector deflects the electron beam in substantially the same direction as the electromagnetic deflector.

The electrostatic deflector preferably operates with an electric field strength that substantially cancels the deflection of the secondary electrons by the electromagnetic deflector.
Alternatively, the ratio between the strength of the deflection electric field by the electrostatic deflector and the strength of the deflection magnetic field by the electromagnetic deflector may be variable.
Further, a means for giving a common electrostatic potential to each electrode of the electrostatic deflector may be further provided.

In the scanning electron microscope included in the present invention, an electrostatic deflector that works in synchronism with the electromagnetic deflector is provided at substantially the same position as the electromagnetic deflector.
In the scanning electron microscope configured as described above, since the secondary electron trajectory is not bent by the deflector, it can be detected efficiently.

  As is clear from the above description, in the charged particle microscope of the present invention, by providing a deflection electric field in addition to the deflection magnetic field, secondary electrons can be detected efficiently even when the deflection region is wide. In addition, the energy region of secondary electrons to be detected can be easily selected.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
(First embodiment)
FIG. 1 is a schematic cross-sectional view of a charged particle microscope according to the first embodiment of the present invention. A focused electron beam is emitted from the electron optical column 101 around a point 113 (referred to as point A) on the wafer 103. Wafer 103 is fixed on stage 109. The stage 109 is placed on the horizontal movement mechanism 104 and the vertical minute movement mechanism 105, and can move in the horizontal direction (xy direction) and the minute distance (Δz) in the vertical direction by these mechanisms. ing.

  These mechanisms are placed on the first support 106, and the tilt angle of the stage 112 around the rotation shaft 111 can be adjusted by the tilt mechanism 107 of the first support 106. The first support 106 is held by the second support 108 via the tilt mechanism 107, and the position of the second support 108 in the vertical direction (z direction) is roughly adjusted by the vertical movement mechanism 109. Is done. Reference numeral 110 denotes a driving device for the vertical movement mechanism 119. The horizontal movement mechanism 104, the vertical minute movement mechanism 105, the tilt mechanism 107, and the driving device 110 are controlled by the stage controller 112. Reference numeral 102 denotes a housing of the entire apparatus.

  A characteristic feature of this embodiment is that a light emitting unit 114 composed of a light source and an illumination optical system, a light receiving unit 115 composed of a light receiving optical system and a photodetector, and a z-axis sensor composed of a height calculation circuit 116 are provided. That is, the light emitting unit 114 and the light receiving unit 115 are attached to a member 108 that supports the stage tilting mechanism 107. The light 117 emitted from the light emitting unit 114 is adjusted so as to be reflected at the beam irradiation position 113 of the wafer 103 and received by the light receiving unit 115. This z-axis sensor is used to measure the height of the irradiation position of light on the sample.

  The irradiation position point A (113) of the charged particle beam is a plane including the rotation axis 111 and is adjusted in advance so as to be in a plane parallel to the moving direction of the stage height fine adjustment mechanism 105. When the point A is deviated from this surface, the position of the point A moves as the stage 112 is inclined. However, it is not particularly difficult to adjust the position of the point A so as to be within an allowable range, for example, about 0.5 μm from the predetermined position.

  This adjustment may be performed as follows, for example. First, the stage is kept horizontal, and a charged particle beam is irradiated to an arbitrary point of interest away from the rotation axis (tilt axis) 111 to obtain a secondary electron image on an image display device (not shown). Watch the movement of the image when the stage is tilted around the horizontal position. If the horizontal plane of the stage does not include the rotation axis, the image moves in one direction when the tilt angle of the stage is changed in one direction before and after the horizontal plane.

  When the height of the stage is changed by the fine adjustment mechanism (vertical minute movement mechanism) 105 and the inclination angle is similarly changed to one direction, it is adjusted to the height at which the image of the attention point turns back at the horizontal position of the stage. . At this time, the height of the irradiation position of the beam coincides with the height (surface) including the rotation axis.

As the fine adjustment mechanism 105, for example, if a piezo element and an enlargement mechanism are used, a resolution of about several nanometers can be easily obtained, which is much smaller than the resolution of the z-axis sensor.
Next, the stage is moved in the horizontal direction, and moved to a position where the movement of the point of interest is minimized when the stage is tilted. At this time, the position of the attention point is substantially on the rotation axis. Thereafter, the electron optical microscope is set so that the irradiation position of the beam comes to this point. Further, the height of the beam irradiation position at this time is measured by a z-axis sensor, and this height is stored in the stage controller 112 as a reference position.

  As described above, the light 117 emitted from the light source in the light emitting unit 114 is imaged at a point 113 (hereinafter referred to as a point A) on the wafer 103 where the charged particle beam is desired to be irradiated by the illumination optical system. The light receiving unit 115 forms an image of the point A on the photodetector. Here, if the height of the point A is slightly changed due to, for example, warpage of the wafer, the movement of the point A is detected by the light receiving unit, and therefore the change in the height of the point A is obtained.

  A change Δh in the height (distance) of the point A from the axis 111 (referred to as axis B) is obtained from the image movement amount and the optical system by an arithmetic circuit, and the value is given to the stage controller 112. In the stage controller 12, the height of the point A is changed by −Δh using the stage height fine adjustment mechanism 105. This places point A on axis B.

  Since the height resolution of a normal z-axis sensor is about 1 μm, the amount of lateral movement of the point A when the stage is tilted is 1 μm or less. This is because, without moving the wafer mechanically using the xy movement mechanism of the stage, the point A is moved to the point of the charged particle beam by moving the scanning range of the charged particle beam within a range where the lens deflection aberration is sufficiently small. This value can be stored in the irradiation area.

  As described above, according to the present invention, even when the height of the wafer surface varies due to wafer warpage or the like, it is possible to minimize the movement of the irradiation position of the charged particle beam when the stage is tilted. Become.

  FIG. 1 also illustrates the second advantage of the present invention. In general, an objective lens of a charged particle optical system has a higher resolution as the distance between the lens and the sample is smaller. However, if the distance between the sample and the lens is narrow, there is a problem that if the sample is tilted greatly, the stage cannot collide because the stage collides with the objective lens. Therefore, the height of the stage is changed according to the required tilt angle and resolution.

  In the present invention, the light emitting portion 114 and the light receiving portion 115 of the z-axis sensor are both attached to the support portion (second support) 108 of the tilt mechanism 107. Therefore, even if the height of the stage is changed, the height relationship between the light emitting unit 114 of the z-axis sensor and the axis B (111) of the light receiving unit 115 does not change. Therefore, it becomes possible to make the height of the point A to be irradiated with charged particles on the wafer coincide with the height of the axis B regardless of the height of the stage.

  Instead of the z-axis sensor that can directly observe the microscope observation point, the height distribution is measured at a place other than the observation point with the sample previously attached to the stage, and the height distribution map is stored. Also good. In this case, the height of the stage at one observation point (first observation point) is adjusted by the method described in the embodiment, and the height from the first observation point based on the height map at other observation points. The stage height is corrected by the difference of.

(Second Embodiment)
FIG. 2 is an overall configuration diagram of a charged particle microscope (scanning electron microscope) according to the second embodiment of the present invention, and FIG. 3 is an enlarged schematic diagram of the periphery of the objective lens. In FIG. 2, reference numeral 125 denotes a field emission type electron gun which generates an electron beam 128 having a very high luminance. The details of the periphery of the electron gun are not described because they are not related to the essence of the invention. The electron beam 128 is further reduced by the condenser lens 126 and the objective lens 127 and irradiated on the surface of the sample 137. At this time, the secondary electron 138 emitted from the sample surface is detected by the detector 133. Reference numeral 136 denotes an aperture.

  The electron beam 128 is scanned two-dimensionally on the sample 137 by the deflector 129. In synchronization with this, the signal of the detector 133 is displayed on the image display device 132, whereby information on the sample surface can be obtained. 131 is a sawtooth voltage generator used for scanning the deflection electrode 129 and the display device 132, and 130 is a power supply for the deflector.

  Further, as a characteristic part of the present invention, an arithmetic circuit 122 to which a disturbance detector 139 is connected and a delay circuit 123 are provided. These functions are described below.

  The electron beam 128 is ideally scanned in a raster shape over the sample 137 as indicated by an arrow 121 in FIG. This scanning is performed by a deflector 129 connected to a sawtooth power source 131. However, if there is an external disturbance (disturbance), such as a magnetic field or vibration, the trajectory will deviate from the original trajectory. 120 is an original trajectory at a certain moment, and N is a position on the sample to be irradiated with the electron beam at that time.

Let the coordinates of N be (x _n , y _n ). Due to the disturbance, the trajectory of the sample is 128, so that R on the sample is irradiated. The coordinates of R are (x _r , y _r ). If the disturbance is known, (x _r , y _r ) can be calculated from (x _n , y _n ).

Consider the case where the disturbance is an external magnetic field. The external magnetic field is detected by a magnetic field detector 139 using, for example, a Hall element placed near the apparatus. (X _r , y _r ) is obtained by the arithmetic circuit 122 from the disturbance information obtained from the detector 139 and (x _n , y _n ).

  This calculation is performed as follows, for example. The relationship between the magnetic field detected by the detector and the irradiation position of the beam is measured in advance using a calibration sample. The irradiation position information of the beam is obtained, for example, by irradiating the fine gold particles formed on the calibration sample with the beam. In many cases, the direction of the magnetic field at the source of the disturbance magnetic field around the device and the direction of the magnetic field at the irradiation position are the same, so there is a high correlation between the measured magnetic field and the deviation of the irradiation position of the beam. , Almost one-to-one correspondence.

From this measurement result, the relationship between the detected magnetic field intensity and the actual beam irradiation position is prepared as a table and stored in a storage circuit in the arithmetic circuit 122. The arithmetic circuit 122 compares the magnetic field obtained from the detector 139 with the above table, and obtains the actual position (x _r , y _r ) of the beam when there is a disturbance by interpolation. A delay time corresponding to the calculation time in the arithmetic circuit 122 is delayed by the delay circuit 123 by the signal from the secondary electron detector 133. The image display device 132 displays the delayed signal using this coordinate information (x _r , y _r ). Thereby, accurate image information can be obtained regardless of disturbance.

(Third embodiment)
FIG. 4 is a diagram showing a schematic configuration of a charged particle microscope according to the third embodiment of the present invention. This embodiment is a modification of the second embodiment, and the same portions are denoted by the same reference numerals, and therefore redundant description is omitted.

  As shown in FIG. 4, two detectors 139a are provided in total at positions orthogonal to each other around the upper portion of the sample 137 to measure the direction and strength of the magnetic field actually felt by the beam. The deviation of the beam irradiation position with respect to the direction and strength of the magnetic field is measured in advance, and the beam irradiation position when there is a disturbance is given based on the measurement result.

  In this way, it is possible to cope with a more accurate case even when there is not one kind of disturbance. Providing a plurality of sets of detectors, for example, two pairs in the lens barrel portion and one pair in the sample chamber is also effective in improving accuracy. In this case, the deflection amount of the trajectory is measured in each part of the lens barrel, and the position for obtaining the beam irradiation position can be obtained as the sum of the deflection amounts in each part.

The above will be described with reference to the optical system shown in FIG. FIG. 5 also shows the optical system from the electron gun 125 to the deflector 129 omitted in FIG. 126 is a condenser lens, and 136 is an aperture. The crossover 135 moves perpendicular to the axis (in the horizontal direction) in the presence of a magnetic field. The value of the disturbance magnetic field between the electron gun 125 and the crossover 135 is measured by the detector pair 139c. Accordingly, the deviation of the crossover position from the signals of the detector pair 139c, for example, d _x1 is obtained as being in the x direction. The method for obtaining d _x1 may be the method described in the example of FIG.

Next, the disturbance magnetic field between the crossover 135 and the lens 127 is determined by the detector pair 139b. The deviation of the trajectory here is effectively given as the movement of the crossover 135 (d _x2 , d _y2 ). When the reduction ratio of the lens 127 is 1 / M, the beam movement on the sample 137 is given by (d _x1 + d _x2 , d _y2 ) / M. However, it is assumed that the influence of image rotation by the lens is included in 1 / M with 1 / M as a matrix.

Finally, the disturbance magnetic field between the lens 127 and the sample 137 is determined by the detector pair 139a. The deviation of the trajectory here is given by (d _x3 , d _y3 ). Therefore, the deviation of the sample irradiation position 124 of the _{beam, ((d x1 + d x2} , d y2) / M + (d x3, d y3) given by. The position where the planned irradiation _{(x n, y n),} the actual in the beam and the position to be irradiated _{(x r, y r),} (x r, y r) = (x n, y n) + (d x1 + d x2, d y2) / M + (d x3, d _y3 ).

  In FIG. 5, the illustration is omitted because the drawing becomes complicated, but the detection signals of the disturbance detectors 139 a, 139 b, and 139 c are respectively input to the arithmetic circuit 122.

  The case where the disturbance is a magnetic field has been described above. A substantially similar method can be applied when the disturbance is vibration. For example, an acceleration pickup is provided in the lens barrel. By measuring the beam irradiation position with respect to the direction and magnitude of the acceleration of the lens barrel vibration when there is floor vibration of a predetermined frequency, a response to the lens barrel vibration at the beam irradiation position is obtained. The displacement is determined as a linear response to acceleration.

For example, in the simplest model that can simply describe the resonance frequency of the relative vibration between the stage and the lens barrel as ω ₀ and the attenuation coefficient as λ, if the response is obtained at at least two different frequencies, the resonance frequency ω ₀ and the attenuation The coefficient λ is obtained. The displacement is determined as a linear response to acceleration. In reality, there are many more complicated cases, so an appropriate model that reproduces the relationship between acceleration and displacement is used. When the system is complicated, the calculation time for obtaining the position of the beam is particularly long. Therefore, the error is large in the conventional correction method, and the effectiveness is further increased by using this method.

(Fourth embodiment)
FIG. 6 is a schematic configuration diagram of a charged particle microscope according to the fourth embodiment of the present invention. The present embodiment is a modification of the second embodiment (or the third embodiment), and the same portions are denoted by the same reference numerals and redundant description is omitted.

  In the present embodiment, means (respectively indicated by reference numerals 134a, 134b, and 134c) for storing the scanning information, the secondary electron detector signal, and the disturbance signal are provided, and the stored disturbance is stored for the stored scanning information. The correction described in the second or third embodiment is performed based on the signal, and an image is displayed together with the stored detection signal. Note that the delay circuit 123 can be omitted or included in the memory circuit 134b.

In this case, the original position information is two-dimensional discrete information. On the other hand, the coordinates after the correction process do not exactly match the discretely given coordinates. A solution to this problem will be described with reference to FIG. It is assumed that the position to which the signal is to be given is given by (Md _x , Nd _y ) (M and N are natural numbers, d _x and _dy are the minimum position steps) and a grid point. For simplicity, d _x and _dy are omitted, and the position is indicated by (M, N). If the corrected coordinates are C (1.3, 3.8), for example, there is no corresponding grid point. In this case, for the four points (1, 3), (2, 3), (1, 4), (2, 4) around this point, the signal corresponding to this point is Distribute according to the positional relationship. Usually, this distribution is performed according to the current of the beam passing through the region represented by each lattice point.

For simplicity, the current distribution of the electron beam is one side, assuming it is the uniform distribution of the square _{d x,} when the _{d x = d y, (1,3} ), (2,3), ( The signal amounts to be allocated to 1, 4) and (2, 4) are distributed to 0.21 vs. 0.09 vs. 0.49 vs. 0.21, respectively.

  Here, the distribution is made to 4 points. However, when the beam spread is larger than the interval of the lattice points, the signal is distributed to the surrounding lattice points of the surrounding 4 points. In reality, the beam distribution is not uniform, for example, it is often a distribution close to a normal distribution. In this case as well, it can be easily understood that the current passing through each lattice point may be obtained and distributed accordingly.

Current distribution of the beam is exp - a proportional to ^{((r / a) 2/} 2). However, a is a beam radius and is r ² = x ² + y ² . The function erf (x) is defined as follows.

  The ratio of the signal amount to each section is given by the following equation.

(Erf (1.3d _y /a)−erf(0.3d _y /a))×(erf(0.8d _x /a)+erf(0.2d _x / a):
(Erf (1.3d _y /a)−erf(0.3d _y /a))×(erf(1.2d _x /a)+erf(0.2d _x / a):
_{(Erf (0.3d y /a)+erf(0.7d y /a))×(erf(0.8d} x /a)+erf(0.2d x / a):
_{(Erf (0.3d y /a)+erf(0.7d y /a))×(erf(1.2d} x /a)-erf(0.2d x / a)
As can be easily understood, other areas are also allocated. For example, in the area of (1, 2), (erf (2.3d _y /a)−erf(1.3d _y /a))×(erf(0.8d _x /a)+erf(0.2d _x / A).

In this case, the signal weight may vary depending on the position. For example, if the position (x ₁ , y ₁ ) is not irradiated with electrons even once due to disturbance, and (x ₂ , y ₂ ) is irradiated multiple times, the apparent (x ₁ , y ₁ ) is It is dark and (x ₂ , y ₂ ) is displayed brightly.

  Normally, if the scanning period of the electron beam and the period of disturbance do not match, the weight at each position is averaged by adding the signals at each point in multiple scans. Can be avoided.

  Furthermore, it is also possible to do the following. That is, a weight is given to each display position based on the scanning information, and the signal amount is normalized by this weight. Specifically, as described above, when a signal is distributed to each grid point, the integral amount of the current applied to the region represented by each grid point, that is, the charge amount is stored in advance, and this charge amount is stored. The signal may be normalized by

  When there is a fast signal, the beam may move across a plurality of regions within the time when one signal position is designated. In this case, a signal for passing through each region is added while weighting a plurality of regions through which the beam has passed. Actually, this time is divided into small times, and the average position is used at each time.

  In addition, when there is a slow and large disturbance, it is also effective to correct the electron beam trajectory itself according to the disturbance in combination with the correction method of the present invention as seen in the prior art. This is because, by correcting the electron beam trajectory itself, the deviation of the beam is considerably reduced, and the correction range of the correction method according to the present invention can be reduced. In this case, of course, the corrected trajectory information is used as the original position information.

  By the way, the electron microscope has been described as an example so far, but the observation principle of the microscope using ions is the same as that of the electron microscope, and therefore, it can be applied in exactly the same manner. In addition, the present invention is not limited to the detection signal of secondary electrons, but can be applied to SIMS analysis for measuring secondary ions, SNMS for measuring secondary neutral particles, or an apparatus for measuring X-rays and light. Absent.

(Fifth embodiment)
FIG. 8 is a schematic cross-sectional view around the objective lens of a charged particle microscope (scanning electron microscope) according to the fifth embodiment of the present invention. The present embodiment relates to an in-lens or semi-in-lens electron microscope, and the arrangement of the electron gun, condenser lens, and objective lens is the same as in FIG. 14, but the position of the secondary electron detector and the position of the deflector Characterized by position and configuration.

  That is, as shown in FIG. 8, the opening part of the pole piece 152 faces downward. The left side of the figure shows an outline of the magnetic field strength distribution along the central axis of the lens. Since a magnetic field is applied to the surface of the sample 146, the trajectory of the secondary electrons 151 is limited to the vicinity of the center of the lens.

  The detector 158 includes a combination of an electron capture bias grid 154, a scintillator 155, and a photomultiplier tube 156. A shielding tube 157 is provided so as to surround the electron beam 142 so that the influence of the bias does not reach the electron beam. Also in this method, the electromagnetic deflector 144 is provided downstream of the detector 158 in order to increase the deflection region of the electron beam.

  A characteristic point of this embodiment is that an electrostatic deflector 161 is provided inside the electromagnetic deflector 144. The combination of the electromagnetic deflector and the electrostatic deflector is, for example, as shown in FIG. 144a to 144d are coils of an electromagnetic deflector, and 145 is a core. Reference numerals 161a to 161d denote electrodes of the electrostatic deflector.

  Here, a current is passed through the coils 144a and 144b in the direction indicated by the arrows, and a negative potential is applied to the electrode 161a and a positive potential is applied to the electrode 161c. Thus, the electric field 222 and the magnetic field 159 are orthogonal to each other in the vicinity of the central axis. FIG. 10 conceptually shows the trajectories of primary electrons and secondary electrons at this time.

Now, the intensity E of the deflection electric field on the central axis, the magnitude of the deflection magnetic field B, and the action length L are both assumed. The energy of primary electrons is U ₁ and the energy of secondary electrons is U ₂ . The electron charge is e, and the mass is m. The primary electron deflection angle is approximately
The amount due to the magnetic field is eBL / (2 mU ₁ ) ^1/2
The amount due to current is eEL / 2U ₁
Total eBL / (2 mU ₁ ) ^1/2 + eEL / 2U ₁
Given in.

On the other hand, the deflection angle of secondary electrons is eBL / ( ₂ mU ₂ ) ^1/2 -eEL / 2U _1.
Given in. Here, it is approximated that the action length of the electric and magnetic fields is the same as that of the primary electrons.

Now, when the deflection electric field E is selected to be E = B (2U ₂ / m) ^1/2 , the deflection angle of secondary electrons whose energy is U ₂ becomes 0. That is, it is not affected at all by the deflector and is detected efficiently. FIG. 10 schematically shows these states. 151a is the electron orbital energy _{U 2,} 151b indicates a trajectory of the time trajectory at high energy than _{U 2,} 151c has energy than _{U 2} conversely low. Therefore, the value of the electric field E is adjusted mainly according to the energy of the secondary electrons to be detected.

  Conversely, the energy region of the secondary electrons to be detected can be selected by adjusting the ratio between the deflection electric field and the deflection magnetic field. For example, it is convenient when a high energy reflected electron image is desired in addition to a low energy secondary electron image.

  When secondary electrons are to be detected in a wide energy range, a positive bias potential may be applied to the deflection electrodes in common. FIG. 11 is a circuit diagram showing a method of applying current or potential to the electromagnetic coil and the deflection electrode. In the example of FIG. 11B, a bias is applied to the deflection electrode by the power source 227. The above explanation is for the case where the bias of the power source 227 is zero.

  For example, if the potential of the deflection electrode is 3 kV in common, the orbit of electrons from several eV to several hundred eV is corrected by the electrostatic deflector regardless of the energy. In this case, as shown in FIG. 10, the axial symmetry of the accelerating electric field by the deflecting electrode can be improved by making the electrodes axisymmetric.

  In the above description, for the sake of simplicity, the action lengths of the deflection electric field and the deflection magnetic field are equal. However, in practice, for example, the required deflection can be achieved by increasing the action length of the deflection electric field by elongating the electrodes. It is possible to reduce the strength of the electric field. As is clear from the above equation, the deflection electric field and the deflection magnetic field are in a proportional relationship. This relationship is easy to realize.

  FIG. 11 (a) shows one example of a circuit that implements the present invention. A current source 223 is connected to the electromagnetic deflector coils 144a and 144b and the electrostatic deflector electrodes 161a and 161c. In series with these, variable resistors 224a and 224b are connected. When a current is passed through the coils 144a and 144c, the potentials of the electrodes 161a and 161c change in proportion to the current to generate a deflection electric field. The ratio between the deflection electric field and the deflection magnetic field can be adjusted by changing the resistances of the variable resistors 224a and 224b. By constructing exactly the same circuit, the magnetic field and electric field of the electromagnetic deflector coils 144b and 144d and the electrostatic deflector electrodes 161d and 161c can be changed in proportion.

  In addition, as shown in FIG. 11B, a signal from the same signal generation circuit 224 can be added to the current amplification circuit 225 and the voltage amplification circuit 226 to generate a synchronized deflection electric field and deflection magnetic field. It is also easy to give an electrostatic potential equally to each electrostatic deflector.

  By the way, although the method for realizing the present invention has been described using examples, it goes without saying that the electric circuit for applying a current or a potential to each coil and deflector is not limited to that illustrated in FIG. Absent.

  Further, when a conductor is used as the electrode material, an eddy current is induced in the electrode when the frequency of electromagnetic deflection is high, and in some cases, the influence of a new deflection due to a magnetic field caused by the eddy current cannot be ignored. In that case, the influence of eddy current can be suppressed by using a high resistance material such as silicon carbide as the electrode material.

  It is also possible to shift the position of the electrostatic electrode and place it in a place where the deflection magnetic field is small. However, since the orbit is rotated by the magnetic field of the lens, the direction of deflection by electrostatic deflection is determined so as to obtain a predetermined deflection angle. This is a simple calculation in terms of electron optics.

1 is a cross-sectional view showing a schematic configuration of a charged particle microscope system according to a first embodiment. The figure which shows schematic structure of the charged particle microscope concerning 2nd Embodiment. The figure which shows typically the structure of the objective lens periphery in 2nd Embodiment. The figure which shows typically the structure of the objective lens periphery of the charged particle microscope concerning 3rd Embodiment. The figure which shows schematic structure of the modification of 3rd Embodiment. The figure which shows schematic structure of the charged particle microscope concerning 4th Embodiment. The figure for demonstrating the distribution method of the signal in 4th Embodiment. The figure which shows schematic structure of the objective lens periphery of the charged particle microscope concerning 5th Embodiment. The figure which shows the example of arrangement | positioning of the electromagnetic deflector and electrostatic deflector in 5th Embodiment. The figure which shows the orbit of the primary electron and secondary electron in 5th Embodiment. The circuit diagram for supplying an electric current or an electric potential to the electromagnetic deflector and electrostatic deflector in 5th Embodiment. Sectional drawing of the conventional charged particle microscope system. The figure for demonstrating the problem which arises in tilting a sample stage in the conventional charged particle microscope system. The figure which shows schematic structure of the conventional charged particle microscope. The figure which shows schematic structure of the objective lens periphery of the conventional charged particle microscope of an in-lens system. The figure which shows the trajectory of the primary electron and the secondary electron in the conventional in-lens type charged particle microscope.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 ... Electro-optic lens tube 102 ... Housing 103 ... Wafer (sample)
DESCRIPTION OF SYMBOLS 104 ... Horizontal moving mechanism 105 ... Vertical minute moving mechanism 106 ... 1st support body 107 ... Inclination mechanism 108 ... 2nd support body 109 ... Sample stage 110 ... Vertical movement mechanism drive device 111 ... Rotation axis (inclination axis)
DESCRIPTION OF SYMBOLS 112 ... Stage controller 113 ... Beam irradiation position 114 ... Light emission part 115 ... Light reception part 116 ... Height arithmetic circuit 117 ... Light beam 119 ... Vertical movement mechanism

Claims (3)

In a charged particle microscope in which a secondary electron signal obtained by scanning an electron beam on a sample with an objective deflector is detected by a detector and displayed as an image,
The detector is disposed upstream of the electron beam from the objective deflector, and the objective deflector includes an electromagnetic deflector and an electrostatic deflector that operates in synchronization with the electromagnetic deflector, and A charged particle microscope characterized in that an electric field of a deflector deflects the electron beam in substantially the same direction as the electromagnetic deflector.   The charged particle microscope according to claim 1, wherein the electrostatic deflector operates with an electric field strength that substantially cancels the deflection of the secondary electrons by the electromagnetic deflector.   3. The charged particle microscope according to claim 1, wherein a ratio between a strength of a deflection electric field by the electrostatic deflector and a strength of a deflection magnetic field by the electromagnetic deflector is variable.

Images