JP7394599B2 - Electron beam inspection equipment using retarding voltage - Google Patents

Description

The present invention relates to an electron beam inspection device using retarding voltage.

Semiconductor devices are shrinking every year in accordance with Moore's law, and the minimum feature size of cutting-edge devices is now 20 nm or less. Achieving small feature sizes requires exposure technology that can form smaller patterns.

Conventionally, laser beams with a wavelength of 193 nm have been used for exposure, but as this greatly exceeds the optical resolution, in recent years, exposure technology that uses EUV light with a wavelength of 13.5 nm has been actively promoted. With the arrival of the Logic 7 nano generation, TSMC and others were the first in the world to successfully put it into practical use.

Evolution of exposure technology means a reduction in the feature size of photomask patterns and an increase in the number of patterns. To perform semiconductor exposure, an insulating quartz plate called a photomask with a pattern drawn thereon is used. A detailed pattern is drawn on this quartz plate based on circuit design data. An original plate called a template for nanoimprinting is also made from quartz.

In order to check whether the patterns formed on these masks have the correct dimensions, electron microscopy technology is essential because optical technology lacks resolution. In order to obtain high-definition electron beam images, it is necessary to irradiate the sample surface with electrons with a small beam spot size and detect signal electrons. It is essential that there is no negative impact on the

Conventionally, retarding methods have been used for the purpose of inexpensively reducing aberrations in electron beam optical systems. In this method, the energy of an electron beam emitted from an electron gun is accelerated to a high energy of several tens of kilovolts, the electron beam is focused using an electron optical system, and a reverse bias voltage is applied to the sample. In this method, the energy of the sample is decelerated to 1 KV or less and then irradiated onto the sample. It is characterized by the ability to maintain high resolution despite the low energy when irradiating the sample.

The higher the energy of an electron beam, the shorter the wavelength, which allows the beam size to be narrowed down and improves image resolution. However, when a sample such as a semiconductor device is directly scanned with a high-energy electron beam to obtain an image and observed, the high-energy electrons can cause semi-permanent effects on the semiconductor sample and damage the device. I've come to understand. In order to prevent this, a retarding method is used in semiconductor CDSEMs to narrow down the beam size and lower the electron beam energy during irradiation.

The retarding method requires that the potential at the measurement point on the sample surface be a desired potential. However, with photomasks or nanoimprint templates made of quartz, which is a dielectric material, unlike semiconductor wafers, which are usually conductive, voltage is applied to the electrodes provided at the bottom of the sample. Since the voltage and the potential on the surface of the photomask are different, it is extremely difficult to always adjust the surface potential to the desired surface potential. Furthermore, when the sample moves due to the movement of the XY stage, the capacitance of the measurement system changes accordingly, resulting in a change in the retarding voltage that is substantially applied to the sample surface.

Furthermore, in devices such as multi-beam inspection devices that form images by accelerating secondary electrons generated on the sample surface to a predetermined energy, there is a problem that the retarding voltage becomes unstable, causing problems in image formation. was there.

The present invention has the above-mentioned conventional feature that it is possible to stably apply a desired retarding voltage to the surface of a sample in which a pattern is formed on an insulating substrate such as a photomask, and obtain a high-resolution and stable image. There is.

In order to solve the above problems, the present invention provides an electron beam inspection apparatus that applies a retarding voltage to a sample and irradiates the sample with a high-speed electron beam at a low speed to generate a high-resolution image. an electron gun that generates an electron beam, an objective lens that focuses the electron beam generated by the electron gun with a predetermined high acceleration voltage onto the sample, and a two-dimensional scan of the electron beam narrowed by the objective lens over the surface of the sample. a retarding voltage application device that applies a predetermined retarding voltage to the sample; a detection device that detects the potential or capacitance of the surface of the sample and is provided at the tip of the objective lens facing the sample; and a correction device that corrects the first retarding voltage so as to correct the potential of the surface of the sample to a constant value based on the potential or capacitance of the surface of the sample detected by the device.

At this time, a sample surface potential control device is provided that controls the second potential on the surface of the sample by applying a voltage externally, and instructions are given to the sample surface potential control device based on the potential or capacitance detected by the detector. Then, the second potential is controlled, or the correcting means is instructed to control the first retarding voltage, or both of them are controlled.

In addition, the coordinates corresponding to the measurement points or the surface of the sample are divided vertically and horizontally, and the surface of the sample is divided into vertical and horizontal directions, and the first The retarding voltage, the second potential, or both are corrected.

In addition, a table is created in which the surface of the sample is divided into vertical and horizontal directions, and pre-measured potentials or capacitances are registered in correspondence with partial areas of each intersection. The first retarding voltage, the second potential, or both are corrected.

In addition, a groove is provided in a band shape around the patterned area on the surface of the sample, and claw electrodes are provided to electrically connect the inner pattern and the outer pattern of the black border, which is electrically isolated. The potential of the pattern can be controlled externally.

Furthermore, a conductive connecting portion is formed on the substrate instead of the claw electrode.

Moreover, the sample is made to form a pattern on an insulating substrate.

The present invention can stably apply a desired retarding voltage to the surface of a sample having a pattern formed on an insulating substrate such as a photomask, thereby obtaining a high-resolution and stable image.

Furthermore, by measuring the potential or capacitance of the sample surface in real time and controlling the retarding voltage or the potential of the sample surface, stable and high-resolution images can always be obtained even if the sample moves.

In addition, by dividing the sample vertically and horizontally and registering the potential or capacitance on the surface of the sample in the table in advance, the retarding voltage and potential on the surface of the sample can be automatically corrected as the sample moves. This allows stable and high-resolution images to be obtained at all times.

FIG. 1 shows a configuration diagram of one embodiment of the present invention.

In FIG. 1, an electron gun 1 is a known electron gun that generates electrons, such as TFE, LaB6, a field emitter, a photoelectron gun, or the like. After adjusting through a focusing lens and aperture (not shown) so that the accelerating voltage is on the order of several hundred to several tens of KV and the current is, for example, on the order of several pA to several hundred nA, it is After narrowing down to a diameter on the order of several hundred nm, the surface of sample 9 is irradiated. A bias voltage (retarding voltage 8) is applied to the substrate (mask) of the sample 9 to optimize the irradiation energy or the energy of electrons generated on the surface of the sample 9 (about several hundred V to 1 KV).

The blanking device 2 turns on and off the electrons generated by the electron gun 1 at high speed.

The blanking aperture 3 is a diaphragm that blocks the electrons deflected by the blanking device 2.

The objective aperture 4 limits electrons entering the objective lens 5.

The objective lens 5 is used to narrow down the electron beam and irradiate the sample 9 with the electron beam.

The deflection device 6 is a two-stage deflection system (electrostatic, electromagnetic) that deflects the electron beam and scans the sample 9 with the narrowly focused electron beam two-dimensionally.

The surface potential/capacitance detection device 7 is a disc-shaped plate with a hole in the center installed at the tip of the objective lens 5, and is used to measure the potential on the surface of the sample 9 and the capacitance (capacitance). This is an electrode for measuring potential or capacitance (with a device not shown). The electrode shape does not have to be circular, but may have other shapes. It doesn't necessarily have to be in the center, nor does it need a hole. The metal itself constituting the objective lens can also be used as the capacitance measuring electrode.

The potential/capacitance detection signal 71 is a signal (surface potential signal or capacitance detection signal) detected by the surface potential/capacitance detection device 7, and the surface potential of the sample 9 is determined by a personal computer (not shown) based on the signal. Alternatively, it is for detecting capacity in real time. In the case of a surface potential signal, it is a signal for measuring the potential of the surface of the sample 9 (or the retarding upper electrode 81), and is a minute potential corresponding to the potential of the surface of the sample 9 (or the retarding upper electrode 81). This is a well-known method that generates a voltage, amplifies it, and measures (detects) its potential. In the case of the capacitance detection signal, it is a known signal for calculating the capacitance by supplying a minute high frequency voltage and detecting the current flowing at that time. In addition, surface potential and capacitance measuring means known in the world can be used.

The retarding voltage 8 is a retarding voltage applied to the sample 9, and is a deceleration voltage (this In an example, the deceleration voltage is 500V or 1KV lower than 5KV to 35KV).

The retarding upper electrode 81 is an electrode for extracting the potential on the upper surface of the sample 9 to the outside.

The retarding lower electrode 82 is an electrode for extracting the potential of the lower surface of the sample 9 to the outside. A retarding voltage 8 is normally applied to this retarding lower electrode 82 .

The insulator 10 electrically insulates the retarding lower electrode 82.

The sample 9 is a sample such as a mask, and is fixed on an insulator 10 and a retarding voltage 8 is applied to the sample 9. The electron beam decelerated by the retarding voltage 8 is applied to the sample 9. The sample 9 is then irradiated with a low-energy electron beam, which is focused to a minute spot using a high voltage through the objective lens 5 and then decelerated using the retarding voltage 8 to reduce damage and contamination. , and can generate high-resolution images.

The XYZ stage 11 is a stage that fixes the sample 9 and automatically moves it to arbitrary coordinates (X, Y) and further to (Z). The movement can be performed while accurately measuring in real time in the X, Y, and, if necessary, Z directions using a laser interferometer (not shown). More specifically, there is an insulator on the stage, and on top of that there is a photomask holding mechanism called a mask pallet made of a conductor, which can also serve as a lower electrode. On the other hand, a claw electrode is separately placed on the pallet, and a voltage can be applied from there to the top of the photomask.

The vacuum chamber 12 is a container that houses the XYZ stage 11 on which the sample 9 is fixed, etc. in a vacuum.

The vacuum pump 13 is an oil-less evacuation system that evacuates the vacuum chamber 12, such as a dry pump or TMP.

The vibration isolator 14 is for vibration isolation so that vibrations are not transmitted to the vacuum chamber 12 or the like from the outside.

The electron detection device 21 is a known detector that detects secondary electrons emitted when the sample 9 is irradiated with an electron beam and scanned two-dimensionally.

Next, the operation of the configuration shown in FIG. 1 will be briefly explained.
(1) A sample 9 such as a mask (a mask fixed to a holder) is automatically transported and fixed to the XYZ stage 11 by a robot.
(2) The position of the photomask is measured in real time using a laser interferometer according to the design data of the pattern provided on the photomask, and the XYZ stage 11 is moved to the specified position, and a predetermined position of the sample 9 is irradiated with an electron beam. At the same time, the surface potential/capacitance detection device 7 measures the potential (or capacitance) of the surface of the sample 9 (or the retarding upper electrode 81) in real time, and it becomes the reference value (or the value of the reference point). (correct one or both of the retarding voltage 8 and the potential of the retarding upper electrode 81) so that the potential on the surface of the sample 9 is always constant even when the sample 9 is moved. As a result, even if the sample 9 is moved, the surface of the sample 9 always has the same potential, and the sample 9 is scanned while being irradiated with a low-acceleration electron beam that is decelerated by the retarding voltage 8. This makes it possible to constantly obtain electronic images and the like. A detailed explanation will be given below.

FIG. 2 shows an example of a sample (mask) of the present invention. FIG. 2 shows the cross-sectional structure of a typical photomask. FIG. 2(a) shows a structural example of an EUV mask, and FIG. 2(b) shows a structural example of a DUV mask.

In FIG. 2A, the EUV mask is generally a Ta-based mask in which a conductive reflective film 32 such as MoSi is provided on ultra-low thermal expansion quartz as an insulating substrate 31, and a pattern 33 is formed thereon. Since the absorbing layer is provided, the area where the pattern is formed is characterized by electrical conductivity throughout.

In FIG. 2(b), in the case of the illustrated DUV mask, there may be no conductive film under the layer of the pattern 34, so the pattern 34 is often a floating island that is not electrically connected anywhere. condition may occur. When such a mask is irradiated with an electron beam, the surface potential changes due to the charge, making it impossible to obtain a stable and high-resolution secondary electron image.

FIG. 3 shows an example of a black border of the present invention. FIG. 3(a) shows a side view, and FIG. 3(b) shows a top view.

In FIG. 3 , photomask 35 is an example of sample 9 .

The bias electrode 36 is an electrode for applying a potential to the back surface of the photomask 35.

The black border 37 is provided to reduce the influence of flare generated by the EUV exposure apparatus, and has a structure in which the enclosed area is electrically isolated from the surrounding area.

The claw electrode 38 is an electrically connected electrode for applying the bias voltage 1 to the surface of the photomask 35 . Although the shape of the electrode is arbitrary, it can also be a needle-shaped electrode in order to break through the insulator on the surface.

Bias voltage 1 is a voltage applied to claw electrode 38 (surface of photomask 35).

Bias voltage 2 is a voltage applied to the back surface of photomask 35.

Next, the configuration will be explained.
(1) In FIG. 3, as described above, the EUV mask in FIG. It is made up of layers. If this continues, the entire photomask surface can be brought to the desired potential by applying a voltage to the peripheral area from the outside.
(2) However, because EUV exposure equipment uses multiple mirrors with a short wavelength of 13.5 nm and the light source is not necessarily high-performance, a problem called flare occurs, causing light to scatter in directions other than the intended direction. Light source quality deteriorates. Therefore, the EUV light also reaches adjacent areas that should not be exposed, causing problems. In order to prevent this, an electrically insulating boundary area called a black border 37 is provided as shown in the figure to surround the area where the main pattern of the photomask is written.

Since the boundary called the black border 37 is formed by carving the MoSi conductor forming the reflective layer down to the quartz glass substrate, the area surrounded by the black border 37 is electrically separated from the surrounding area and becomes an insulator.

If the mask becomes insulated, the region surrounded by the insulator will not reach the desired potential even if a voltage is applied from around the mask surface. Furthermore, when observing by irradiating an electron beam such as with an SEM, electric charge gradually accumulates at the observation point, resulting in deterioration of the image. In order to avoid this, the present invention provides a special claw electrode 38 for energizing the area surrounded by the black border 37 as shown in FIG. 3. As a result, conduction occurs between the periphery and the inside of the mask, allowing charge to escape and making it possible to maintain the same surface potential. The claw electrode 38 used at this time has a structure as thin and flat as possible, and its surface is coated with an insulator with high dielectric strength such as Teflon (registered trademark). The distance between the two is approximately several mm (for example, 3 mm) to prevent discharge from occurring. Not only one claw electrode 38 but also a plurality of claw electrodes 38 may be provided around the photomask. Furthermore, since there may be a plurality of insulating regions separated by black borders, the voltages applied to the plurality of claw electrodes 38 may be independently controlled. Furthermore, it is desirable to be able to apply independent voltages inside and outside the black border.

FIG. 4 shows an example of a black border of the present invention.

In FIG. 4, the black border 41 is formed by carving the MoSi conductor forming the reflective layer into the quartz glass substrate, as described above.

The peripheral region 42 is a peripheral region electrically separated by the black border 41.

The measurement target area 43 is a measurement target area separated by a black border 41.

The connecting portion 44 is a conductive portion formed to electrically connect the measurement target region 43 and the peripheral region 42 electrically separated by the black border 41.

Next, FIG. 4 will be explained in detail.
(1) In FIG. 4, the black border itself has a very large size compared to the pattern size actually formed. Therefore, a pattern (connecting portion 44) is formed in which the periphery and the inner region are connected somewhere, leaving a portion of the border to such an extent that the performance of the black border 41 is not limited. In this way, electrical continuity occurs between the peripheral part of the mask partitioned by the black border 41 and the inside, so that the potentials of the two regions can be made the same. Furthermore, it becomes possible to release charges generated by electron beam irradiation, and image distortion due to charge-up phenomenon is eliminated. It is sufficient to apply voltage using the claw electrodes to the area around the outer periphery of the mask.
(2) On the other hand, it may not always be possible to form the above pattern (connecting portion 44). At that time, conduction can be established by forming wiring on the black border 41 using a conductive material such as Ta or W that has the property of absorbing EUV light using an FIB used for mask modification, an electron beam mask modification device, or the like. Multiple conduction points can be taken. Since the mask correction device can also form wiring three-dimensionally, wiring across the black border 41 is easy. As a result, the area partitioned by the black border 41 is connected to the outer periphery, and conduction can be established by contacting the outer periphery with an electrode from the outside.The mask surface has the desired potential everywhere, and the EUV mask is charged. You can prevent it from uploading.

FIG. 5 shows a schematic illustration of the surface potential of the present invention.

In FIG. 5, C1 represents the capacitance between the lower surface of the objective lens 5 and the surface of the mask (retarding upper electrode 81).

C2 represents the capacitance between the mask surface (retarding upper electrode 81) and the bias electrode (retarding lower electrode 82).

E represents the retarding voltage 8.

Next, the operation will be explained.
(1) In Fig. 5, this is a schematic representation of the principle that determines the potential generated on the internal area partitioned in an insulated state by a black border placed inside the electron beam inspection equipment, or on the surface of a DUV photomask in a floating state. It is. A photomask does not necessarily have a conductive reflective layer like an EUV mask, so the entire surface is not necessarily electrically conductive. In that case, when an external voltage is applied to the floating region from the bottom electrode of the substrate, the surface potential is equal to the capacitance. Determined. Opposing the photomask surface (the surface of sample 9), there is an objective lens 5 made of metal inside the electron beam inspection apparatus chamber. Since the photomask surface and the objective lens 5 are close to each other, it is equivalent to the presence of two metal electrodes, and a large electric capacitance C1 is formed. On the other hand, since a dielectric material such as quartz exists between the front surface of the photomask and the back surface of the photomask, a large capacitance C2 is formed between the pattern on the front surface of the photomask and the lower electrode 82 of the photomask.

Therefore, the potential of the photomask surface located in the middle is approximately determined by the ratio of the sizes of the two capacitances.
(2) When the photomask is moved by the XYZ stage 11 during measurement, the distance to the objective lens 5 and the distance from peripheral members change, so the capacitance C1 of the measurement system described above changes. The bias voltage (retarding voltage 8) applied to the substrate is distributed as this changes, so if the same voltage is applied, the surface potential changes as the XYZ stage 11 moves. In other words, if a measurement is performed with a constant retarding voltage applied, the bias potential generated on the substrate surface differs depending on the location, resulting in a completely different measurement being taken. In the worst case, the image may not be formed or the position where the image is formed may fluctuate, making measurement unstable.
(3) Therefore, the potential of the bias electrode (retarding voltage 8) or the potential of the mask surface is adjusted so that the potential of the mask surface (potential distributed between capacitance C1 and capacitance C2) is always constant even if the XYZ stage 11 moves. If the potential (first potential) is corrected and automatically controlled so that the potential on the mask surface remains constant, the potential on the mask surface will remain constant even if the XYZ stage is moved, resulting in a stable and high-resolution secondary It becomes possible to acquire electronic images.

FIG. 6 shows an example of a capacitance/potential map table of the mask of the present invention.

FIG. 6(a) shows an example of a capacitance map table, and FIG. 6(b) shows an example of a surface potential map table.

(a) of FIG. 6 shows that the XYZ stage 11 (or mask) is divided into i parts in the X direction and j parts in the Y direction, and the capacitance C1 (i, j) at the intersection position (i, j) is calculated. It was measured and registered.

(b) of FIG. 6 shows the potential V(i, j) at the intersection position (i, j) of the XYZ stage 11 (or mask) divided into i parts in the X direction and j parts in the Y direction. It was measured and registered.

Here, the capacitance C1 and the potential V are measured as follows.
(1) Sample surface potential V: There are various methods of determining the sample surface potential V. Surface potential can be determined using commercially available surface electrometers with various principles and types. A surface electrometer can be placed at the tip of the objective lens 5 to measure the potential at the measurement location on the mask. The voltage applied to the substrate (the retarding voltage 8 in FIG. 1 or the potential of the retarding upper electrode 81) while the XYZ stage 11 is moving or stopped is automatically adjusted so that the measured potential becomes a constant value.
(2) Similar results can be obtained by measuring the capacitance generated between the objective lens and the sample surface. The capacitance is measured at a certain coordinate on the XY coordinates. The sample 9 is moved by the XYZ stage 11 and the capacitance is measured. Since the capacitance difference represents a change in surface potential, the voltage applied to the sample 9 (the retarding voltage 8 in FIG. 1 or the potential of the retarding upper electrode 81) is changed by the amount of the change in capacitance, so that the surface potential becomes substantially the desired value. Adjust so that it is.
(3) The surface potential V can also be determined by actually irradiating an electron beam and examining the energy of secondary electrons generated. Specifically, the energy of secondary electrons can be investigated by installing a mesh grid etc. on the surface of the secondary electron detection device, applying a predetermined voltage, and examining the amount of secondary electrons detected after passing through the mesh. . By changing the grid voltage, the energy distribution can be determined, and as a result, changes in the sample surface potential V, which is the source of the energy shift of secondary electrons, can be detected. The surface potential can be maintained at a desired value by automatically adjusting the bias voltage applied to the substrate (the retarding voltage 8 in FIG. 1 or the potential of the retarding upper electrode 81) so that there is no change in the measured value.
(4) The surface potential change or capacitance change resulting from the movement of the XYZ stage 11 determined as described above is an amount specific to a specific device, and remains constant unless the inside of the device is changed.

Therefore, once data on changes in surface potential V or changes in capacitance due to XYZ stage movement is acquired and registered, the amount of correction can be estimated by inputting position coordinates. Therefore, data of surface potential or capacitance change corresponding to such measurement point coordinates XY is stored in a storage device as a map table as shown in (b) and (a) of FIG. 6, and the map table is used during actual measurement. By automatically adjusting the voltage applied to the mask (sample 9) based on , the surface potential can be kept constant at a desired value.

FIG. 7 shows a flowchart for setting the mask surface potential of the present invention.

In FIG. 7, S1 moves the XY stage. Here, it is assumed that the coordinates of the movement position are (Xi, Yj).

S2 measures surface potential or capacitance. This can be done by measuring the potential Vij on the surface of the mask facing the objective lens 5 while moving the XY stage to coordinates (i, j), or by measuring the capacitance Cij between the objective lens 5 and the surface of the mask facing the objective lens 5. Measure.

S3 calculates a correction value. In this process, a correction value is calculated based on the surface potential Vij or capacitance Cij measured in S2 so as to match the reference value, the potential (capacitance) at a specific location, or an arbitrary specified value.

S4 sets the applied voltage. This is done by setting the potential of the retarding lower electrode 82 or the retarding upper electrode 81 in FIG. 1 to a predetermined value, and correcting it so that the surface potential or capacitance is always constant.

In S5, it is determined whether the measurement is completed. If YES, the process ends. If NO, repeat S1 and subsequent steps.

As described above, by measuring the surface potential or capacitance every time the XYZ stage 11 is moved and correcting it so that it is always constant (correcting the potential of the retarding lower electrode 82 or the retarding upper electrode 81 in FIG. 1), The surface potential of the mask (sample 9) can be kept constant, making it possible to obtain stable and high-resolution secondary electron images. When correcting in real time, the surface potential and capacitance can be measured at any measurement point and the applied retarding voltage value can be corrected.

FIG. 8 shows a mask surface potential setting flowchart (Part 2) of the present invention. 8 shows the coordinates of the mask from the (b) surface potential map table in FIG. 6 or (a) capacitance map table in FIG. 6, which are measured and registered in advance, instead of measuring the surface potential and capacitance in real time as in FIG. The surface potential and capacitance were read out and corrected.

In FIG. 8, S11 moves the XY stage. Here, the moving position is assumed to be coordinates (Xi, Yj).

S12 reads the correction value. As described above, the surface potential or capacitance at the coordinates of the mask is read from the surface potential map table (b) in FIG. 6 or the capacitance map table (a) in FIG. 6 which are registered in advance. If there is no matching coordinate in the table, read out the values (potential or capacitance) of multiple coordinates in the vicinity.

S13 calculates a correction value. This calculates a correction value based on the surface potential Vij or capacitance Cij read in S12 so as to match the reference value, the potential (capacitance) at a specific location, or an arbitrary specified value.

S14 sets the applied voltage. This is done by setting the potential of the retarding lower electrode 82 or the retarding upper electrode 81 in FIG. 1 to a predetermined value, and correcting it so that the surface potential or capacitance is always constant.

In S15, it is determined whether the measurement is completed. If YES, the process ends. If NO, repeat S11 and subsequent steps.

As described above, each time the XYZ stage 11 is moved, the surface potential or capacitance is read out from the table registered in advance and corrected so that it is always constant (the potential of the retarding lower electrode 82 or the retarding upper electrode 81 in FIG. 1 is corrected). By doing so, the surface potential of the mask (sample 9) can be kept constant, and a stable and high-resolution secondary electron image can be obtained. Note that since the Z-axis height affects the capacitance value, it is desirable to automatically control the height to a desired constant height (distance between the sample surface and the objective lens).

FIG. 9 shows a configuration diagram of another embodiment of the present invention. FIG. 9 shows an external view of the multi-electron beam inspection device. The multi-electron beam inspection device irradiates the sample 9 with multiple electron beams at once in a planar manner, re-accelerates the secondary electrons generated in the sample 9, and forms an image, thereby achieving ultra-high-speed two-dimensional image acquisition. This is a device that can do this. In this device, it is very important to uniformly accelerate the secondary electrons generated on the surface of the sample 9. If the surface potential varies, the operation of the beam splitter 541 will be impaired and an image will not be focused on the desired position. Control of potential is very important.

In general, inspection devices perform high-speed, large-volume measurements, so they may move step-by-step like a normal SEM or review device, or the XYZ stage 11 may move continuously. The present invention can be used even in such cases. Specifically, it is as follows.
(1) In the case of step-by-step, it is sufficient to have the correction values for each step as a map (see FIG. 6). On the other hand, when the stage 11 moves continuously, the following procedure is performed. For example, as shown in FIG. 6, the area of the photomask is divided into about 100 equal parts, and the surface potential or capacitance of the sample 9 at the center coordinates of each area is measured. It is stored as a map (see FIG. 6). Since the surface potential or capacitance value has the property of changing gently as the photomask moves, correction values at coordinates that cannot be directly measured can be estimated by interpolation.
(2) For example, using the measurement point coordinates and map data output from the laser interferometer, interpolation calculations such as least squares method, spline, Lagrangian interpolation, etc., or Gaussian filter can be used for intermediate coordinates that are not in the table in Figure 6. It is possible to estimate the surface potential or capacitance at the measurement point coordinates by performing two-dimensional weighting calculations such as the following. A correction value at any coordinate can be estimated from this value. When the stage is continuously moved, the XYZ coordinates of the measurement point are captured and correction is performed in real time.
(3) On the other hand, the global retarding potential applied to the measurement target area is determined by the above-mentioned principle, but when considering local capacitance changes, it is necessary to The surface potential can be estimated by calculating the area of the pattern that contributes to capacitance formation directly under the objective lens 55 using pattern design data (CAD data such as GDI), and calculating the capacitance. By calculating and controlling the correction amount from the capacitance estimated in this way (controlling the potential of the retarding upper electrode 81 or the retarding lower electrode 82), the potential of the measurement point on the surface of the sample 9 can be set to a desired value. I can do it.

FIG. 1 is a configuration diagram of an embodiment of the present invention. This is an example of a sample (mask) of the present invention. It is an example of the black border of this invention. It is an example of the black border of this invention. FIG. 3 is a schematic explanatory diagram of the mask surface potential of the present invention. It is an example of a map table of capacitance and potential of the mask of the present invention. 3 is a flowchart for setting a mask surface potential according to the present invention. FIG. 2 is a flowchart (Part 2) for setting the mask surface potential of the present invention. FIG. FIG. 3 is a configuration diagram of another embodiment of the present invention.

1: Electron gun 2: Blanking device 3: Blanking aperture 4: Objective aperture 5: Objective lens 6: Deflection device 7: Surface potential/capacitance detection device 71: Potential/capacitance detection signal 8: Retarding voltage 81: Retarding Upper electrode 82: Retarding lower electrode 83: First potential 9: Sample 10: Insulator 11: XYZ stage 12: Vacuum chamber 13: Vacuum pump 14: Vibration isolator 21: Electron emission transport 31: Insulating substrate 32: Conductive reflective layer 33: absorption layer (pattern)
34: Pattern 35: Photomask 36: Bias electrode 37: Black border 38: Claw electrode 41: Black border 42: Peripheral area 43: Measurement target area 44: Connection part

Claims (5)

In an electron beam inspection system that applies a retarding voltage to a sample and irradiates it with a high-speed electron beam at a low speed to generate a high-resolution image,
an electron gun that generates an electron beam with high acceleration voltage;
an objective lens that narrows and irradiates the sample with a high acceleration voltage electron beam generated by the electron gun;
a deflection system that two-dimensionally scans the surface of the sample with an electron beam narrowed by the objective lens;
a retarding voltage application device that applies a retarding voltage to the sample;
a detection device that detects the potential or capacitance of the surface of the sample and is disposed at the tip of the objective lens facing the sample to detect the potential or capacitance of the surface of the measurement point of the sample;
a correction device that corrects the retarding voltage based on the potential or capacitance of the surface of the sample detected by the detection device so that the potential of the surface of the sample is constant;
A groove is provided in a belt shape around the patterned area on the surface of the sample to electrically connect the pattern inside the black border and the surrounding area , and the potential of the internal pattern is applied from the outside. An electron beam inspection device using a retarding voltage, comprising a claw electrode for controlling, or a conductive connecting portion provided on the substrate of the sample for controlling the potential of the internal pattern from the outside. A sample surface potential control device is provided to control the potential of the surface of the sample by applying a voltage from the outside, and based on the potential or capacitance detected by the detection device, instructions are given to the sample surface potential control device to 2. The electron beam inspection apparatus using a retarding voltage according to claim 1, wherein the retarding voltage is controlled by controlling the potential on the surface of the sample, or by instructing the correction device to control the retarding voltage, or both. The surface of the sample is divided into vertical and horizontal directions, and the retarding voltage or the voltage of the sample is adjusted according to the movement of the sample, based on the potential or capacitance measured in real time by dividing the surface of the sample into partial areas at each intersection point. 3. An electron beam inspection device using a retarding voltage according to claim 1, wherein the surface potential or both are corrected. The surface of the sample is divided vertically and horizontally, and a table is provided in which pre-measured potentials or capacitances are registered in correspondence with partial areas of each intersection point, and based on the table, the voltage is adjusted according to the movement of the sample. 3. The electron beam inspection apparatus using a retarding voltage according to claim 1, wherein the retarding voltage, the potential of the surface of the sample, or both are corrected. 5. The electron beam inspection apparatus using a retarding voltage according to claim 4, wherein the sample has a pattern formed on an insulating substrate.

Images