JP2000331635A - Scanning electron microscope - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress the electrification of the surface while keeping the resolution to reduce the distortion of a secondary electron image by positive electrification by adding the function of modulating the scanning speed of electron or the luminance of an electron beam. to interrupt the scanning of the electron beam. SOLUTION: An electron beam 110 incident on a silicon wafer 111 by an electron gun 101 is scanned by a horizontal scanning electromagnetic coil 102 and a vertical scanning electromagnetic coil 103. A vertical scanning electromagnetic coil current 105 has a saw-tooth waveform, while a horizontal scanning electromagnetic coil current 104 is modulated to a step-like waveform by a horizontal scanning signal modulation device 107. Consequently, a locus 112 of the electron beam for scanning the electron beam on the surface of the silicon wafer 111 is formed of the part horizontally rested or slowly scanned and the part scanned at high speed. This repeat of scanning is set to about 11-10 Hz. Since the amplitude of scanning is changed with a constant scanning period when the observation magnification is changed, the linear speed of scanning is changed inversely proportional to the magnification.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a scanning electron microscope used for shape observation and dimension control in a semiconductor manufacturing process.

[0002]

2. Description of the Related Art In a scanning electron microscope used for shape observation and dimension control in a semiconductor manufacturing process, the processing dimensions of an insulating material such as a silicon oxide film, a silicon nitride film, and a photoresist, which are easily charged, are accurately measured. Is required. In the current mainstream system, the influence of charging is avoided by setting the energy of primary electrons incident on the sample to 1.5 keV or less. Hereinafter, the reason will be described.

An electron has a negative charge, and when a primary electron enters, a secondary electron having a current comparable to the primary electron current is emitted from the surface, and the surface charge is deeply related to the secondary electron emission phenomenon. I have. In fact, conventional antistatic methods
This property of secondary electron emission is used. Curve 4 in FIG.
01 schematically shows the relationship between the energy of the electron beam and the secondary electron emission ratio on the sample surface. The secondary electron emission ratio is a value obtained by dividing the current of secondary electrons by the current of primary electrons.
For example, when primary electrons are vertically incident on a silicon oxide film, the secondary electron emission ratio gradually increases as the energy of the primary electrons increases from a low energy to 100 eV.
Before and after, the secondary electron emission ratio exceeds 1 (E1 in FIG. 4). When the energy of the primary electrons is further increased, the secondary electron emission ratio reaches a maximum value at around 300 eV (Ema in FIG. 4).
x). Further, the energy of the primary electrons is increased to about 1.
When it exceeds 5 keV, the secondary electron emission ratio again falls below 1 (E2 in FIG. 4). The values of E1, E2, and Emax vary depending on the material. As described above, when the secondary electron emission ratio is smaller than 1, the surface of the substance is negatively charged. When the secondary electron emission ratio is larger than 1, the surface of the substance is positively charged. That is, since the secondary electron emission ratio changes according to the energy of the primary electrons, the charging mode changes according to the energy of the primary electrons.

FIGS. 5A and 5B schematically show the relationship between the secondary electron emission ratio and the charged state of the insulating film surface. In the figure, 501 is a primary electron, 502 is a high energy component of a secondary electron, 503 is a low energy component of a secondary electron,
504 is a secondary electron, and 505 is an insulating sample. FIG.
FIG. 4A shows the case where the secondary electron emission ratio is larger than 1, and FIG. 4 corresponds to the case where the primary electron energy is between E1 and E2. When the secondary electron emission ratio is larger than 1, electrons are emitted more from the surface, and the surface is positively charged.
On the other hand, FIG. 5B shows the case where the secondary electron emission ratio is smaller than 1, and FIG. 4 corresponds to the case where the primary electron energy is higher than E2. When the secondary electron emission ratio is less than 1, electrons are more likely to remain on the surface, and the surface is negatively charged.

In fact, at the boundary of the secondary electron emission ratio of 1 as described above, not only the polarity of the charge is reversed, but also a large difference described below occurs between the two. That is, when the energy of the primary electrons is sufficiently large and the secondary electron emission ratio is smaller than 1 and negatively charged, the surface is negatively charged,
The primary electrons will be decelerated. Therefore, if there is no other leakage or the like, the charging voltage of the primary electrons is reduced to the energy of E2, and the charging voltage proceeds until the secondary electron emission ratio approaches 1. Therefore, the charging voltage is the difference between E2 and the primary electron energy, and it is not uncommon for the charging voltage to exceed -100V. When such charging occurs, the secondary electron image is greatly distorted, and the length measurement error increases.

On the other hand, when the secondary electron emission ratio is larger than 1, the surface is positively charged, but when charged several volts, the low-energy component of secondary electrons having energy of only several eV is drawn back to the surface. Immediately, the incident current of the primary electrons 501 and the secondary electrons 503 pulled back to the surface balances the emission current of the emitted secondary electrons 502, and charging does not proceed further.

As described above, when the secondary electron emission ratio is lower than 1, a large negative charge is generated, and when the secondary electron emission ratio is larger than 1, the charge falls within the positive charge of several volts.

For the above reasons, the energy of the primary electrons of the scanning electron microscope used for shape observation and dimension control in the semiconductor manufacturing process is selected in the range from E1 to E2 in FIG. The value of E2 is in the range of 500 eV to several keV for the main material, and the energy of primary electrons is actually in the range of 500 eV to 1.5 keV.

By the way, in a scanning electron microscope, an electromagnetic lens is generally adopted to obtain a practical focal length. However, since this electromagnetic lens utilizes Lorentz force acting on electrons, it is in principle used on the low energy side. It has the property that chromatic aberration (energy aberration) is significantly increased. Since the minimum width of the energy width of the electron beam is determined by the characteristics of the electron source, the resolution is maintained from 500 eV to 1.
In order to realize a relatively low primary electron energy of 5 keV, it is necessary to take measures to prevent the chromatic aberration of the electromagnetic lens constituting the electron optical system from increasing due to the low energy.
As this method, generally, a method is employed in which the primary electron beam is maintained at a high energy, passes through an electromagnetic lens system, and is decelerated immediately before the sample. FIG. 3 shows a typical structure of a conventional scanning electron microscope. 3, reference numeral 301 denotes a vacuum vessel, 302 denotes an electron source, 303 denotes primary electrons, 304
Is a sample, 305 is a secondary electron, 306 is a secondary electron detector,
307 is a condenser lens, 308 is an axis adjustment coil,
309 is a stigma adjustment coil, 310 is a scanning coil,
311 is an objective lens, 312 is a sample stage, 313 is an electron acceleration power supply, and 314 is an electronic deceleration power supply. With such a configuration, the energy of the primary electrons is reduced while maintaining the resolution.

FIG. 2 shows an electron beam scanning method of a conventional scanning electron microscope. In FIG. 2, 21 is an electron gun, 22 is a horizontal scanning electromagnetic coil, 23 is a vertical scanning electromagnetic coil, and 24 is a horizontal scanning electromagnetic coil. An electromagnetic coil current, 25 is a vertical scanning electromagnetic coil current, 26 is an electron beam locus, 27 is a detector, 28 is a silicon wafer, and 29 is an electron beam.

Although not shown in detail in FIG.
Is composed of a horizontal scanning electromagnetic coil 22 and a vertical scanning electromagnetic coil 23 for deflecting and scanning an electron beam, a cathode for emitting electrons, an electrode system for accelerating electrons, and an electromagnetic lens system for converging electrons. ing.

The scanning electron microscope is a horizontal scanning electromagnetic coil 22.
A secondary electron generated by an electron beam 29 scanned on a trajectory 26 on a silicon wafer 28 by a vertical scanning electromagnetic coil 23 is detected by a detector 27, and a change in a signal due to a surface shape or material is synchronized with a scanning position. Then, the image is formed, and the shape and the like are observed. To change the magnification, change the scanning area. The trajectory for obtaining the secondary electron image in the trajectory 26 is indicated by a solid line, and the trajectory not related to the image is indicated by a broken line. As can be easily understood from the drawings, the conventional method employs a method of irradiating a continuous trajectory with an electron beam on the entire surface of a portion to be observed.

Usually, by appropriately designing the electron optical system, the obtained electron beam can be narrowed down to a beam diameter of several nm, and the secondary electron image can have a spatial resolution of several nm. . Therefore, it is a technique suitable for managing fine shapes and dimensions in a semiconductor manufacturing process and the like, and is currently widely used.

[0014]

However, in the conventional configuration, an electron optical system of an acceleration / deceleration system is employed in order to realize the low energy of the primary electrons while maintaining the resolution. There is a new problem that the positive charge increases.

FIG. 6 shows that the primary electron energy is constant (1.
1 keV), a negative bias was applied to the sample experimentally, and the change in the surface potential at that time was examined. However, the bias voltage is subtracted. The observation magnification was 100 times (602 in FIG. 6) and 2000 times (603 in FIG. 6). As can be seen, when the deceleration bias is increased and the observation magnification is reduced, the charge amount increases synergistically, and its absolute value is, for example, 100 times and the bias is -3.
When it is set to 00V, it has reached + 30V.

That is, the positive charge in the conventional configuration is
It is related to the observation magnification and the deceleration electric field. The mechanism of this positive charge increase can be understood by considering the three-dimensional effect of seepage of the electric field from the charged portion, as described below.

When charging occurs on a continuous surface, the electric field exudes to a distance approximately equal to the size of the charged surface. The charging of an insulating film on a conductive substrate such as silicon can be approximately regarded as an electric double layer, and it is well known that the exudation of an electric field is about the same as the size of the surface. Nature.

Here, the nature of the electric double layer will be briefly supplemented. The electric potential formed by the electric double layer is represented by a surface potential of V
When s (V) and a solid angle in which the surface of the electric double layer is viewed from the observation point is Ω (sr), it is given by the following equation. V = Vs × Ω / 2π (1) The solid angle Ω included in the equation (1) is determined by the ratio of the distance between the observation point and the spread of the surface forming the electric double layer, and decreases with the distance from the surface. . Therefore, the decreasing speed is faster as the spread of the surface forming the electric double layer is smaller. In other words, the seepage of the electric field becomes shorter as the spread of the surface forming the electric double layer becomes smaller. In FIG. 8,
The difference in the distribution of potential from the surface according to the charging pattern, that is, the shape of the electric double layer is compared. It can be seen that the potential approaches 0 with the distance from the surface. The details of the difference between the patterns will be described in the section for solving the problem, and thus the description is omitted here.

As described above, the larger the scanning area is, that is, the lower the observation magnification is, the longer the exudation distance of the electric field is. In the absence of an external electric field, the recharge of the low-speed secondary electrons is determined by the charging voltage on the surface. Therefore, it is considered that the charging voltage is not affected even if the scanning area, that is, the seeping distance of the electric field changes. However, when an external electric field is applied, the effect that the external electric field is superimposed on the exuded electric field changes depending on the distance of the exudation, so that the charging voltage is affected.

For example, in FIG. 6, the scanning range at the time of observation of 100 times is about 1 mm, and the electric field (not including the influence of charging) applied to the surface at a bias of -300 V is about 20 V / mm. there were. In this case, a potential difference of 20 V is newly generated by an external electric field of 20 V / mm over a distance of 1 mm where the electric field seeps. The potential difference newly generated by the external electric field promotes emission of secondary electrons from the surface. Therefore, the balance of the current balance by drawing back the low-energy secondary electrons is more positively charged than when there is no external electric field (+30 in FIG. 6).
V) state. On the other hand, when the observation magnification is increased from 100 times to 2000 times, the scanning width becomes 1/20 (the scanning area becomes 1/4).
00), the distance over which the electric field seeps is reduced to 1/20, and the potential difference generated by the external electric field within that distance is also reduced to 1/2.
0 becomes smaller. Therefore, the deviation of the equilibrium state caused by the external electric field is smaller than in the case of low magnification. In FIG. 6, the difference from 100 times to 2000 times falls to about 4 times, but this is because the seepage of the electric field is not a simple relation such as a proportional relation to the distance.

From the above considerations, it can be understood why the positive charge is increased by lowering the observation magnification or increasing the deceleration electric field. In other words, when observing a continuous portion having a large surface area, the positive charge increases due to the influence of the external electric field.

An object of the present invention is to suppress surface charging while maintaining resolution, and to reduce distortion of a secondary electron image due to positive charging.

[0023]

As described above, when observing a continuous portion having a large surface area, positive charging increases due to the influence of an external electric field. It is sufficient to scan the electron beam so that observation at low magnification is performed not on a continuous surface but on a discontinuous discrete surface.

The scanning electron microscope of the present invention has a configuration in which the function of intermittently scanning the electron beam is added to the conventional scanning electron microscope by modulating the electron scanning speed or the electron beam luminance. . Furthermore, since the influence of charging is reduced during high-magnification measurement, the mode is automatically switched to continuous irradiation in order to give priority to image resolution.

By the means for interrupting the scanning of the electron beam, a portion of the observation region which is strongly irradiated with the electron beam is divided into discrete regions smaller than the observation region. For example, a length of 100 μm is divided into ten, and high-speed and low-speed scanning is repeated every 10 μm. In this manner, scanning is performed in a checkered manner at a low speed. Then, the size of the electric double layer is about one section (10 μm).

FIG. 8 (b) shows the result of calculation to estimate how the seepage of the electric field changes according to the various scanning patterns shown in FIG. 8 (a). The vertical axis indicates the potential normalized by the surface potential. And the horizontal axis represents the distance from the surface. In FIG. 8A, a region (pattern) 801 represents a square-charged pattern having a side length of 10 (in any unit), and a region (pattern) 802 has a length of one side. Represents a 1 (same unit as the pattern 801) square charged pattern, and a region (pattern) 803
Is a checkerboard pattern with a side length of 1 (pattern 801) in a square with a side length of 10 (the same unit as pattern 801).
(The same unit as the above)). Next, curve 804 plots the potential formed by the large pattern of pattern 801 as a function of distance from the center of the large square surface, and curve 805 is formed by the small pattern of pattern 802. Potential is plotted as a function of distance from the center of the surface of the small square, curve 8
06 is a plot of the potential formed by the checkerboard pattern of pattern 803 as a function of distance from the center of the charged small square surface. In addition,
The potentials are all normalized by the surface potential.
Take the value of When it is desired to know the actual potential, the normalized potential is multiplied by the surface potential.

The potential 804 formed by the large pattern 801 decreases to about 30% at a distance 10 which is the same as the length of one side. In addition, small pattern 8
The potential 805 formed by 02 also decreases to about 30% at the same distance 1 as the length of one side. On the other hand, the potential 8 formed by the checkered pattern 803
In the area 06, the potential decreases rapidly as in the case of the small pattern 803 where the distance from the surface is small, and in the area where the distance from the surface is large the potential 807 changes in the same manner as the potential 807 which is approximately half the potential 804 due to the large pattern 801. ing.

The reason why the potential distribution of the checkerboard pattern 803 changes as described above with the distance from the surface is explained as follows. That is, while the distance from the surface is short, the solid angle for observing the charged surface is determined by each small square pattern as in the case of the small square 802, whereas when the distance is long, the entire charged area becomes larger. Contributes to the solid angle, and approaches the pattern charging of the large square 801, but since it is divided in a checkered pattern, the total solid angle of them is also about half, so the potential formed by 801 Approaching about half the curve.

Next, as a method of intermittently scanning, there are a method of modulating the scanning speed and a method of modulating the electron beam luminance. The method of modulating the luminance makes it possible to intermittently interrupt the electron beam. However, the method of modulating the scanning speed does not completely interrupt the electron beam. Note that modulating the scanning speed has the advantage of requiring less modification from a conventional scanning electron microscope. In order to realize intermittent in an equivalent sense by a method of modulating the scanning speed, the modulation ratio of the speed is important as described below.

First, the relationship between the amount of charge in one scan by an electron beam and the scanning speed will be considered. For simplicity, it is assumed that the current of the electron beam is I (A), the lateral spread is uniform within a square of side a (m), and the electron beam is scanned in the side direction at a speed v (m / s). . Secondary electron emission ratio is σ (where σ> 1)
The thickness of the insulating film on the silicon wafer being observed is d (m), and the dielectric constant is ε.

The charge density Qo left on the surface by the scanned electrons
(C / m ² ) means that the area through which the electron beam passes per unit time is a
× v (m ² / s), while the sum of the incident electron current and the secondary electron emission is (σ−1) × I (A), so that Qo = (σ−1) × I / ( a × v) (2) On the other hand, the capacitance Co per unit area of the surface (F /
m ² ) is given by Co = ε / d (3) Therefore, the charging voltage V (V) in question is: V = Qo / Co = (σ−1) × I × d / (a × ε × v) (4) Therefore, if the electron current and its current distribution are constant, and the film thickness, dielectric constant, and secondary electron emission ratio of the insulating film portion of the sample being observed are constant, the charging voltage in one scan is equal to the scan voltage. It can be seen that it is inversely proportional to speed.

Next, changing the viewpoint, the speed of progress of charging in the scanning area is considered. Here, it is assumed that the time required for one scan among the repetitive scans is sufficiently shorter than the time to be noticed, and the electron beam is repeatedly scanned at a constant speed. In this case, it can be considered that an electron beam whose total current is equal to the primary electron current and whose current density is uniform within the scanning range is incident. Therefore, the total current flowing into the scanning range when the secondary electrons are also taken into account is represented by I
Assuming that t, the area of the scanning region is S (m ² ), and the capacitance of the area of the scanning range is C (F), the rate of voltage increase per unit time R (V / s) is R = It / C = (σ-1) × I / (Co × S) (5) As an insulating film (for example, a silicon oxide film) on a silicon wafer, the relative dielectric constant is around 3.9 and the film thickness is 1 μm.
In this case, the value of C in equation (3) is 35 (μF / m
² ) Now, assuming that the electron current is 10 pA, 10
Assuming that a region of 0 μm square (corresponding to a capacitance of 0.35 pF) is repeatedly scanned with an electron beam, and assuming a secondary electron emission ratio of 1.5, the rate of increase R in voltage per unit time is about 14 V / s. Therefore, for example, +100
The time required to charge to V is 7 s. This 7s
Is about the same as the time (several seconds) required to accumulate the signal from the detector and obtain a secondary electron image. Therefore,
For example, if the scanning speed is increased by a factor of 10, the charging voltage at that portion will fall within 10 V for the same elapsed time of 7 s. Therefore, it can be seen that by locally increasing the scanning speed, the effect of suppressing the progress of charging of the portion where the scanning speed is increased within the time for obtaining the secondary electron image can be sufficiently expected.

The above discussion varies depending on the object to be measured and the magnification. However, in practice, it is sufficient that the charging of the high-speed scanning portion is negligible. We think that 100 is enough.

Therefore, according to this method, a scanning electron microscope in which distortion of an image due to surface charging is extremely small can be realized.

[0035]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment will be described below with reference to the drawings.

FIG. 1 shows the configuration and operation of a scanning electron microscope according to the present invention. In FIG. 1, 101 is an electron gun, 102 is a horizontal scanning electromagnetic coil, 103 is a vertical scanning electromagnetic coil, and 104 is a horizontal scanning electromagnetic coil. Electromagnetic coil current, 105: vertical scanning electromagnetic coil current, 106: scanning coil driving power supply, 107: horizontal scanning signal modulation device, 108: control device, 109: display device, 110: electron beam, 111: silicon wafer, 112: scanning Line 113 is a detector. In 112, the portion where the scanning time is short is particularly indicated by a broken line.

The operation of the scanning electron microscope configured as described above will be described below. First, the electron gun 101
Beam 110 incident on a silicon wafer 111 due to
Is a horizontal scanning electromagnetic coil 1 similar to a conventional scanning electron microscope.
02 is scanned by the vertical scanning electromagnetic coil 103.
The vertical scanning electromagnetic coil current 105 for vertical scanning is a sawtooth wave similar to the conventional one, but the horizontal scanning electromagnetic coil current 104 for horizontal scanning is the same as the horizontal scanning signal modulator 107.
Is modulated into a step-like waveform. As a result, the range 11 for scanning the electron beam on the surface of the silicon wafer 111 is obtained.
Reference numeral 2 is a repetition of a portion that is scanned or stopped slowly (solid line) in the horizontal direction and a portion that is scanned at high speed (dashed line). The repetition of scanning is repeated at about 1 to 10 Hz. Usually, when the observation magnification is changed, the scanning amplitude is changed while keeping the scanning period constant, so that the scanning linear velocity changes in inverse proportion to the magnification.

Now, a square visual field with a side of 100 μm is set to 10
Observing at a magnification of 00, and assuming that the number of scanning lines is 500, the average linear velocity of the portion to be scanned is about 100 cm / s when scanning is repeated at 10 Hz. Assuming that the division ratio between the high-speed scanning portion and the low-speed scanning portion is 1: 1 and the ratio between the linear speed of the high-speed scanning portion and the low-speed scanning portion is 10, the linear speed is 550 cm / s in the high-speed portion and 55 cm in the low-speed portion. / S.

It should be noted that in the drawing, the range 112 for scanning the electron beam is shown.
Is written to the same extent as the size of the silicon wafer 111, but this is for the sake of making the scanning lines easier to see. Actually, while the semiconductor wafer has a diameter of several tens of cm, the observation magnification is, for example, The width of the scanning range when the magnification is 1000 times is about 100 μm.

Therefore, by the above-described operation, a portion where the amount of irradiation of the electron beam is large becomes a portion which is discretely separated within the scanning range. Therefore, even if those portions are charged, the range of the electric field is limited to the size of the discrete portions, and the influence on the observed image is extremely small.

On the other hand, when the observation magnification is increased and the scanning range is reduced, the influence of charging does not reach far from the surface, so that there is no need to interrupt the electron beam. Rather,
Continuous scanning improves the image quality without skipping data. In order to avoid this problem, the present invention has a function of transmitting a control signal for operating the horizontal scanning signal modulation device 107 from the control device 108 only when the observation magnification is less than 10,000 times, for example.

By modulating the scanning speed by the above method, the scanning can be pseudo-interrupted and the scanning region can be pseudo-divided into small regions separated from each other. In the present embodiment, as shown in FIG. 7A, the pattern of the low-speed portion, which is the pseudo-scanning area, has a checkered pattern. In FIG. 7, reference numeral 701 denotes a scanning area (in a simulated sense);
02 is a non-scanning area (in a simulated sense). By making the scanning area checkered, a new effect of reducing the occurrence of striation, which is likely to occur when observing a periodic pattern, can be expected.

As described above, according to the present embodiment, by providing the function of modulating the speed of electron beam scanning according to the scanning area of the electron beam, a scanning electron microscope with extremely little image distortion due to charging is realized. it can.

As described with reference to FIG. 8 as a means for solving the problem, the more the checkerboard pattern is subdivided and the smaller the subdivided area, the greater the effect. As can be seen, it goes without saying that the number of divisions in the horizontal or vertical direction may be practically the number of pixels taken in as an image, that is, 256 or 512.

If the size of one side of one square in the case of scanning in a checkered pattern is made smaller than the value obtained by dividing the scanning length by the number of divisions, it can be expected that the effect will be even greater. Of course can be made smaller than the value obtained by dividing the scanning length by the number of divisions. However, at this time, the irradiation amount of electrons per unit area locally increases, and this may cause deterioration of the surface depending on the material to be observed and the degree of vacuum at the time of observation. Therefore, it is necessary to select an appropriate value. There is.

In the present embodiment, the intermittent operation of the electron beam is performed by modulating the scanning speed. However, there is a method of realizing this by luminance modulation.

In this embodiment, a checkered pattern is used as a pattern for intermittently scanning the electron beam. However, as shown in FIG.

Although the checkerboard and the tortoiseshell are described as examples here, it goes without saying that other patterns may be employed.

In the above invention, the observation magnification is 1
It is assumed that a control signal for operating the horizontal scanning signal modulation device 107 is transmitted from the control device 108 only when the magnification is reduced by a factor of 10,000, but the control signal is set so as to perform the same operation at another appropriate magnification, for example, 1000 times. Needless to say, it can be changed.

In the present embodiment, the method of intermittent scanning is a raster scan method. However, it is needless to say that a raster scan method may be used in each divided area.

[0051]

As described above, according to the present invention, by providing a function of intermittently scanning an electron beam, an excellent scanning electron microscope with extremely little image distortion due to surface charging can be realized. In the conventional method, only 10
An error of not less than nm is caused. For example, when the pattern size of the integrated circuit is about 100 nm, it becomes an error factor that greatly affects device characteristics. However, the method of the present invention can provide a scanning electron microscope capable of achieving an error of 1 nm or less and having sufficient performance even in the manufacture of a semiconductor device in which the minimum dimension of a pattern is less than 100 nm.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a scanning electron microscope according to a first embodiment of the present invention.

FIG. 2 is a configuration diagram of a conventional scanning electron microscope.

FIG. 3 is a configuration diagram of a conventional scanning electron microscope.

FIG. 4 is a diagram showing a relationship between a secondary electron emission ratio and primary electron energy.

FIG. 5 is a diagram showing a relationship between a secondary electron emission ratio and charging of an insulator surface.

FIG. 6 is a diagram showing a change in a charging voltage depending on a sample bias voltage and an observation magnification.

FIG. 7 is a diagram showing a method of dividing a scanning area.

FIG. 8 is a diagram showing a comparison of the relationship between the distance from the surface and the potential depending on the shape of charging.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 101 Electron gun 102 Horizontal scanning electromagnetic coil 103 Vertical scanning electromagnetic coil 104 Horizontal scanning electromagnetic coil current 105 Vertical scanning electromagnetic coil current 106 Scanning coil drive power supply 107 Horizontal scanning signal modulation device 108 Control device 109 Display device 110 Electron beam 111 Silicon wafer 112 Electron Line trajectory 113 Detector 701 Scanning area 702 Non-scanning area

Claims (7)

[Claims] 1. A scanning electron microscope for irradiating a sample with an electron beam while scanning it, detecting secondary electrons and observing the shape of the sample, when the scanning length of the electron beam exceeds a predetermined value. A scanning electron microscope comprising a mechanism for automatically functioning the velocity modulation of a scanning electron microscope. 2. The scanning electron microscope according to claim 1, further comprising a function of modulating the scanning speed so that the low-speed portion has a checkered shape within the scanning range. 3. The scanning electron microscope according to claim 1, further comprising a function of modulating a horizontal scanning speed so that the low-speed portion has a turtle-like shape within a scanning range. 4. A scanning electron microscope for irradiating a sample with an electron beam while scanning, detecting secondary electrons and observing the shape of the sample, wherein the scanning length of the electron beam exceeds a predetermined value. A scanning electron microscope comprising a mechanism for automatically functioning the luminance modulation. 5. The scanning electron microscope according to claim 4, further comprising a function of modulating a scanning speed so that the high-brightness portion has a checkered shape within a scanning range. 6. The scanning electron microscope according to claim 4, further comprising a function of modulating a scanning speed so that the high-brightness portion has a turtle-shape within a scanning range. 7. The scanning electron microscope according to claim 1, wherein spot scanning is performed at a low magnification and uniform scanning is performed at a high magnification.