JP2019054007A - Electrostatic chuck mechanism, and charged particle beam device - Google Patents

Abstract

To provide an electrostatic chuck mechanism capable of suppressing various influences of electric fields on a beam that occur near a sample edge, and a charged particle beam device.SOLUTION: An electrostatic chuck mechanism comprises: a sample suction surface (3) for sucking a sample; a first electrode (4) to which a voltage for generating suction force between the sample suction surface and the sample; a second electrode (6) having a first plane (8) below the first electrode; and a negative voltage application power source for applying a negative voltage to the second electrode. The second electrode has a second plane (9) enclosing the first plane, which is the upper most plane of the second electrode positioned below the sample suction surface. There is also provided a charged particle beam device.SELECTED DRAWING: Figure 18

Description

  The present invention relates to an electrostatic chuck mechanism and a charged particle beam apparatus, and more particularly to an electrostatic chuck mechanism and a charged particle beam apparatus that can effectively suppress the influence of an electric field generated near the edge of a sample.

  The miniaturization of semiconductor devices has progressed exponentially, and in recent years, devices made with dimensions on the order of several tens of nm have become mainstream. In a device manufacturing line having such a fine structure, an apparatus using a scanning electron microscope is used to measure a fine pattern and inspect defects on the device. For example, a dimension measuring SEM (Critical-Dimension Scanning Electron Microscope: CD-SEM) is used for dimension measurement of a gate or a contact hole of a semiconductor device, and a defect inspection SEM is used for defect inspection. In addition, a scanning electron microscope is used for continuity inspection of a wiring deep hole by utilizing potential contrast.

  In a charged particle beam apparatus typified by a scanning electron microscope or the like, an electrostatic chuck mechanism may be used as a holding mechanism for holding a target sample (semiconductor wafer or the like). The electrostatic chuck applies a voltage to the metal electrode provided inside, generates positive and negative charges on the surface of the object to be adsorbed and the electrostatic chuck, and fixes the object to be adsorbed by Coulomb force acting between them. It is. On the other hand, when an electrostatic chuck is used as a sample holding mechanism for an electron microscope or the like, a strong electric field formed between the sample and the electrostatic chuck may disturb the potential distribution on the outer periphery of the wafer (sample). Since the electron beam is bent, there is a possibility that the beam cannot be irradiated to an appropriate position near the outer periphery of the wafer. Patent Document 1 discloses a method in which the chucking electrode of the electrostatic chuck is made larger than the wafer, and the electric field in the vicinity of the wafer edge is made uniform by the electric field created by the portion protruding from the wafer. Patent Document 2 describes an electrostatic chuck mechanism in which an electric field correction component having the same height as the sample surface is disposed near the outer periphery of the sample.

Patent No. 5143787 (corresponding US Patent Publication US2012 / 0070066) Japanese Patent Laying-Open No. 2009-302415 (corresponding US Patent Publication US2009 / 0309043)

  According to the electrode having a diameter larger than the outer periphery of the wafer disclosed in Patent Document 1 and the electric field correction component disclosed in Patent Document 2, it is possible to suppress the disturbance of the potential distribution generated in the outer periphery of the wafer. On the other hand, the inventors have clarified that when a sample is mounted on the electrostatic chuck, charging may occur near the edge of the sample due to the contact. Further, it is conceivable that secondary electrons generated when a beam is irradiated near the edge of the sample adhere to the electrostatic chuck. Such charging can also bend the beam.

  Even by applying a voltage to the correction electrodes as described in Patent Documents 1 and 2, the influence of the electric field on the beam can be alleviated to some extent. When these compounds are combined, the effect cannot be sufficiently suppressed.

  In the following, an electrostatic chuck mechanism and a charged particle beam device for the purpose of effectively suppressing the influence of various electric fields generated near the edge of the sample on the beam are proposed.

  As one mode for achieving the above object, there is provided a static chuck having a sample adsorption surface for adsorbing a sample, and a first electrode to which a voltage for generating an adsorption force is applied between the sample adsorption surface and the sample. An electric chuck mechanism, comprising: a second electrode having a first surface below the first electrode; and a negative voltage application power source for applying a negative voltage to the second electrode, wherein the second electrode Includes an electrostatic chuck mechanism including a second surface that surrounds the first surface and is the uppermost surface of the second electrode located below the sample adsorption surface, and a charged particle beam device. suggest.

  According to the above configuration, it is possible to effectively suppress the influence of various electric fields generated near the edge of the sample.

The figure which shows the outline | summary of a scanning electron microscope. The flowchart which shows the measurement process of the semiconductor wafer using a scanning electron microscope. The figure which shows the electric potential distribution of the objective lens vicinity when a retarding voltage is applied. The figure which shows the electric potential distribution of the sample edge vicinity when a retarding voltage is applied. The figure which shows the example which correct | amends the electric field which generate | occur | produces in the sample edge vicinity when a retarding voltage is applied by the voltage application to the adsorption | suction electrode of an electrostatic chuck. The figure which shows a mode that electric charge generate | occur | produced in the adsorption | suction surface of an electrostatic chuck by contact etc. of a sample and an electrostatic chuck. The figure which shows an example of the electrostatic chuck provided with the electrode which correct | amends the electric field produced by making the vicinity of the edge of a sample non-contact with the adsorption surface of an electrostatic chuck, and applying the voltage to an adsorption electrode. The figure which shows another example of the electrostatic chuck provided with the electrode which correct | amends the electric field produced by making the vicinity of the edge of a sample non-contact with the adsorption surface of an electrostatic chuck, and applying the voltage to an adsorption electrode. The figure which shows a mode that the sample has shielded the electric charge which generate | occur | produced on the adsorption surface of the electrostatic chuck. The figure which shows the example which provided the shield electrode in the electrostatic chuck. The figure which shows the example which provided the shield electrode outside the electrostatic chuck. The flowchart which shows the measurement process of the semiconductor wafer using a scanning electron microscope. The flowchart which shows the measurement process of the semiconductor wafer using a scanning electron microscope. The figure which shows the outline | summary of a scanning electron microscope. The flowchart which shows the measurement process of the semiconductor wafer using a scanning electron microscope. The figure which shows the relationship between the voltage applied to the electrode which correct | amends an electric field, and a sample position. The figure which shows the example which provided the level | step difference in the surface of an electrostatic chuck, and had the correction electrode inside the lower level side. The figure explaining the positional relationship of a sample, an electrostatic chuck, and a correction electrode.

  The embodiment described below mainly relates to a scanning electron microscope that performs measurement of line width and hole diameter of a fine pattern in a semiconductor device, defect inspection on the semiconductor device, and image acquisition. First, the basic principle of an electron beam application apparatus such as a scanning electron microscope will be briefly described. Primary electrons are emitted from the electron gun and accelerated by applying a voltage. Thereafter, the beam diameter of the electron beam is narrowed down by an electromagnetic lens. This electron beam is scanned two-dimensionally on a sample such as a semiconductor wafer. Secondary electrons generated when the scanned electron beam enters the sample are detected by a detector. Since the intensity of the secondary electrons reflects the shape of the sample surface, the fine pattern on the sample can be imaged by synchronizing the scanning of the electron beam and the detection of the secondary electrons and displaying them on the monitor. For example, in the CD-SEM, when measuring the line width of the gate electrode, the edge of the pattern is discriminated based on the change in brightness of the obtained image to derive the dimension. In the case of the defect inspection SEM, the defect is recognized from the obtained image and automatically classified.

  Since these electron beam application apparatuses are used for dimension measurement and defect inspection in a semiconductor manufacturing line, not only performance as an electron microscope such as resolution and length measurement reproducibility but also throughput is very important.

  There are a plurality of factors that determine the throughput, but the effects that are particularly significant are the moving speed of the stage on which the wafer is loaded and the time required for autofocus when acquiring an image.

  There is an electrostatic chuck mechanism as a mechanism for holding a semiconductor wafer or the like to be measured or inspected. If the wafer can be stably fixed by the electrostatic chuck, the wafer can be transferred at a high acceleration and a high speed without being displaced from the stage. Also, with an electrostatic chuck, the entire surface of the wafer can be warped with a substantially uniform force, and the wafer can be flattened and fixed, so that the current flowing through the coil of the objective lens to achieve a uniform height distribution within the wafer surface and focus adjustment. The time for determining the value, that is, the autofocus time is shortened.

  In addition, there is an index that represents the observable range, that is, how much range within the wafer surface can be observed, among important performances as a CD-SEM and defect inspection SEM. In the semiconductor manufacturing line, we want to reduce the manufacturing cost by manufacturing as many chips as possible from a single wafer. On the other hand, the observable range is strictly required.

  On the other hand, when the electrostatic chuck is larger than the wafer, charging may be formed on the surface of the electrostatic chuck. Since the electrostatic chuck is made of a ceramic having high electrical insulation, charging may occur due to contact or friction between the wafer and the electrostatic chuck or irradiation of an electron beam. The charge formed near the wafer edge on the electrostatic chuck disturbs the electric field near the wafer edge, which affects the electron beam when observing the vicinity of the edge on the next wafer to be transported, limiting the observable range. May end up.

  In order to suppress the disturbance of the electric field near the wafer edge, it may be possible to use an electrode with an electric field equalization function in addition to the adsorption function, but it is necessary to correct the warped wafer and ensure heat transfer between the wafer and the electrostatic chuck. In some cases, it is difficult to achieve both the original function as an electrostatic chuck. Furthermore, when the electric field on the wafer surface is controlled by the chucking electrode of the electrostatic chuck, the electric field between the wafer back surface and the chucking electrode becomes very strong, so that the function of equalizing the electric field is limited by the limit of the withstand voltage.

  In the embodiment described below, the influence on the electron beam of the electrostatic chuck internal electrode and the electrostatic charge on the electrostatic chuck is suppressed, and it is easy to achieve both the attractive force of the wafer and heat transfer, and is limited by the withstand voltage. We propose electric field uniformity technology that is difficult to receive.

  In the present embodiment, an electrostatic chuck mechanism in which the outer diameter of the electrostatic chuck is made smaller than that of the wafer, a conductive member is provided outside the electrostatic chuck, and an appropriate voltage is applied to the conductive member, and this is used. The charged particle beam apparatus will be described. According to such a configuration, since the outer diameter of the electrostatic chuck is smaller than that of the wafer, the influence of the electrostatic chuck attracting electrode and the influence of the electric field generated by the electrostatic charge on the electrostatic chuck are shielded by the wafer itself. It is possible to ensure a stable observation range. In addition, since the electric field equalizing function is provided independently of the attracting electrode, it is easy to achieve both the attracting force and the heat transfer characteristics, which are the original functions of the electrostatic chuck. Further, since the conductive member can ensure insulation by the vacuum space, it is difficult to be restricted by the withstand voltage.

  Next, a specific example of the electrostatic chuck mechanism will be described with reference to the drawings. An example in which the outer diameter of the electrostatic chuck is larger than the sample will also be described in order to explain the effect of making the outer diameter of the electrostatic chuck (contact surface with the wafer) smaller than that of the wafer.

  1 to 5 are views showing an outline of a charged particle beam apparatus provided with an electrostatic chuck mechanism. Here, a CD-SEM will be described as an example. FIG. 1 shows an overall view of the apparatus. The basic configuration is that a column 101 for controlling an electron beam, a sample chamber 102 including an XY stage 104 on which an electrostatic chuck 105 for holding a sample is mounted, and a preliminary exhaust chamber 103 for performing vacuum evacuation before the wafer is carried into the sample chamber. It is. The sample chamber 102 is evacuated by a vacuum pump (not shown). The CD-SEM is controlled by a control device (not shown), and the voltage applied to the electrostatic chuck and peripheral electrodes is controlled under the conditions described later.

  FIG. 2 is a flowchart showing a measurement process using a CD-SEM. The wafer is first carried into the preliminary evacuation chamber (step 201), evacuated to a predetermined pressure (step 202), and then carried into the sample chamber (step 203). The wafer carried into the sample chamber is placed on the electrostatic chuck (step 204), a predetermined voltage is applied to the internal electrode of the electrostatic chuck (step 205), and the wafer is attracted onto the electrostatic chuck. Subsequently, after a predetermined retarding voltage is applied (step 206), it moves to each measurement point (step 207) and performs image acquisition and length measurement processing (step 208). This is repeated for all the measurement points, and when the processing for all the measurement points is completed, the wafer is unloaded (step 209).

  FIG. 3 shows the state of the electric field near the observation point when observing the wafer inner area. Reference numeral 301 denotes a wafer, 302 denotes an electromagnetic lens for converging an electron beam, and a broken line 303 denotes an equipotential surface. During observation, by applying a retarding voltage to the electrode built in the electrostatic chuck, the wafer 301 becomes a retarding potential pressure, and the space between the opposing electromagnetic lens 302 is an axis relative to the central axis 304 of the electron beam. A symmetrical electric field is formed, and the electron beam reaches the wafer by drawing a trajectory 305 perpendicular to the equipotential surface.

  In the wafer inner region, the electric field from the surroundings is shielded by the wafer itself, and the region through which the electron beam passes is kept axially symmetric. On the other hand, as shown in FIG. 4, when there is no structure around the wafer periphery, for example, when there is no structure around the wafer, a step is generated on the equipotential surface 401 so as to be compared with the step of the wafer itself, so that an electron beam passes through. The axial symmetry of the electric field in the region is broken. At this time, the trajectory 402 of the electron beam is bent due to the collapse of the equipotential surface, so that the position reaching the wafer is deviated from a desired position. If this amount of deviation is large, the desired pattern will deviate from the field of view of the image, making automatic observation difficult. Therefore, the observable region is narrowed to a range where the influence of the step cannot be ignored.

  In order to suppress the influence of such an electric field, as shown in FIG. 5, the outer diameter of the electrostatic chuck 501 is larger than that of the wafer, and the electric field created by the adsorption electrode 502 protruding from the wafer is pushed back by the step, It is conceivable that the electric field 504 in the electron beam trajectory 503 is kept axially symmetrical as far as possible. However, since the surface of the electrostatic chuck is made of highly insulating ceramics, as shown in FIG. 6, a charge 601 is formed by contact with the wafer, friction, and irradiation of an electron beam when the vicinity of the wafer periphery is observed. The

  Here, as an example, the state of the electric field when the electrostatic chuck is positively charged is illustrated. The electric field generated by the charging on the electrostatic chuck further destroys the axial symmetry of the equipotential surface 603 on the electron beam trajectory 602 and narrows the observable range. A method of applying an appropriate voltage in consideration of the electrostatic charge on the electrostatic chuck is also conceivable. However, the charge on the insulator is usually not uniform and occurs extremely locally. Correction is difficult. Therefore, in the electrostatic chuck having an outer diameter larger than that of the wafer, the observable region is narrowed to a range where the influence of charging on the electrostatic chuck can be ignored.

  Therefore, as illustrated in FIG. 7, the influence of charging as described above is suppressed by making the suction surface of the electrostatic chuck 701 with the wafer 702 smaller than the diameter of the wafer 702. The electrostatic chuck 701 illustrated in FIG. 7 has a feature that the outer diameter is smaller than that of the wafer 702, and the influence of charging on the surface of the electrostatic chuck is shielded by the wafer itself. On the other hand, the disturbance of the electric field when observing the vicinity of the wafer periphery is corrected by providing a conductive member 703 outside the wafer and applying a predetermined voltage Vc. The conductive member 703 has a shape that surrounds both the downward direction and the lateral direction of the wafer, and is installed while maintaining a clearance d between the wafer side surface and the height h1 of the conductive member.

  FIG. 18 is a diagram for explaining in more detail the arrangement conditions of the components constituting the electrostatic chuck. The sample adsorption surface 3 of the electrostatic chuck 2 is formed smaller than the contact side surface of the sample 1. The attracting electrode 4 (first electrode) is built in the electrostatic chuck 2, and a voltage is applied to generate a Coulomb force or the like. Although not shown, a retarding voltage for decelerating the electron beam applied to the sample 1 is applied to the adsorption electrode 4. The electrode 6 (second electrode) provided with the first electrode surface 8 at a position relatively separated from the sample adsorption surface 3 in the Z direction and connected to the electric field correction power source 5 is connected to the electrostatic chuck 2. Connected. The Z direction is the same direction as the optical axis of the electron beam. In this case, the sample adsorption surface 3 is an XY plane orthogonal to the Z direction. Further, the electrode 6 is formed so that the first electrode surface 8 is positioned so that the imaginary straight line 7 which is in contact with the edge of the sample 1 and extends in the Z direction passes.

  The reason why the diameter of the sample attracting surface 3 of the electrostatic chuck 2 is made smaller than that of the sample 1 is to suppress charging generated near the edge of the sample 1 as described above. If there is a sample adsorption surface right next to the edge of the sample, there is a possibility that electric charges accumulate due to friction between the sample and the sample adsorption surface. In addition, there is a possibility that charges are accumulated by landing of electrons generated by irradiating a beam near the edge of the sample. In order to suppress such charge accumulation, the size of the sample adsorption surface is set so that the edge portion of the sample (contact surface side of the edge portion with the electrostatic chuck) is not in contact with the electrostatic chuck. To do.

  Further, if the sample adsorption surface is formed small so that the edge portion of the sample is not contacted, the electric field generated by the voltage applied to the adsorption electrode 4 may leak toward the edge side. In order to effectively suppress the influence of the leakage electric field at the edge portion, the virtual straight line 7 passes below the edge of the sample 1 (the virtual straight line 7 passes, and is separated from the sample 1 relative to the sample adsorption surface 3 and the adsorption electrode 4). The electrode 6 having the first electrode surface 8 is installed at the position). According to such electrode arrangement conditions, the leakage electric field toward the edge side of the sample 1 can be suppressed. According to such a configuration, an equipotential line formed with the electrostatic chuck 3 as a base point can be drawn toward the electrode surface 8. That is, the amount of the electric field leaking toward the edge can be suppressed, and as a result, the beam deflection based on the disturbance of the electric field in the edge portion can be suppressed.

  Note that a stepped step may be provided in the conductive member 801 as shown in FIG. In the example of FIG. 8, an electrode surface 801 having a difference of h1 from the sample adsorption surface, an electrode surface 802 having a difference of h2, and an electrode surface 803 having a difference of h3 are formed. As illustrated, the relationship between the differences is h3> h2> h1. With such a structure, it is possible to adjust the contribution ratio of the electric field correction from the lower side and the electric field correction from the side direction by adjusting the height dimension h2 of the step and the gap dimension d in the side direction. This makes it easier to optimize the correction voltage.

  Furthermore, as illustrated in FIGS. 8 and 18, the correction of the electric field can be performed more accurately by configuring the correction electrode so that the height of the electrode surface increases as the distance from the adsorption electrode 4 increases. In the example of FIG. 18, the second electrode surface 9 is formed, and the electrode surface is formed so as to approach the objective lens side as the distance from the adsorption electrode increases. In the examples of FIGS. 8 and 18, the edge of the sample is not in contact with other members, and the equipotential lines that travel on the sample surface move downward (in the direction away from the objective lens) as they move away from the sample in the XY direction. Head for. When the beam is irradiated near the edge of the sample, it is desirable that the equipotential lines are formed in parallel to the sample surface. Therefore, in order to make the equipotential line parallel to the sample surface, as an electrode for forming an equipotential line that goes against the change of the equipotential line before correction, the electrode surface approaches the objective lens as it moves away from the adsorption electrode. Is adopted.

  According to the structure illustrated in FIGS. 8 and 18, the potential gradient at the edge portion can be corrected so as to be parallel to the sample surface. Further, it may be formed in a slope shape instead of a step shape.

  FIG. 9 shows an example of the electric field distribution on the outer periphery of the sample, which is an electrostatic chuck having an attracting surface smaller than the sample, and an electrode for correcting the electric field is provided below the edge of the sample. By applying a predetermined correction voltage Vc to the conductive member 901, the drop of the equipotential surface 903 due to the step of the wafer itself is pushed back, and the electric field in the electron beam trajectory 904 is kept axially symmetric. The charge 905 on the electrostatic chuck is generated in the same manner when the outer diameter of the electrostatic chuck is smaller than that of the wafer. However, the electric field generated by the charge 905 is shielded by the wafer itself, and thus affects the electron beam trajectory. There is nothing. With such a configuration, it is possible to expand the length measurement possible range without being affected by the electrostatic chuck charging. Also, the conductive member 901 to which the correction voltage Vc is applied is not charged because it is grounded via the power source.

  In addition, by positioning the conductive member 901 below the sample adsorption surface, for example, when the sample is transported out of position, the risk of collision between the electrostatic chuck structure and the sample can be reduced. Become. 7 to 9, the height of the conductive member is shown at a position lower than that of the wafer. However, the correction effect may be further enhanced by making it the same height as the wafer or higher than the wafer.

  In addition, the conductive member is not necessarily a separate component from the electrostatic chuck, and the conductive member may be embedded and integrated in the electrostatic chuck. For example, as shown in FIG. 17, a conductive member 1703 is embedded below the outer peripheral portion of the wafer inside the electrostatic chuck while keeping the wafer suction portion 1702 of the electrostatic chuck 1701 smaller than the outer diameter of the wafer. At this time, the area outside the wafer attracting portion 1702 is made lower than the wafer attracting portion, and the distance 1704 between the electrostatic chuck surface and the wafer is kept to such an extent that the influence of charging can be ignored. With such a configuration, it becomes possible to perform electron beam trajectory correction without being affected by the electrostatic chuck surface charging. In this embodiment, the inner diameter of the conductive member is made smaller than the outer diameter of the wafer. The inner diameter of the conductive member may be larger than the outer diameter of the wafer. However, when the inner diameter of the conductive member is smaller than the outer diameter of the wafer as in this embodiment, the wafer is transferred as shown in the fourth embodiment. A voltage application sequence that takes into account variations and the gap between the electrode and the wafer is not necessary, and the electron trajectory correction becomes easier.

  Further, if only the edge portion of the sample is made non-contact, it is not necessary to provide a difference in height between the suction electrode and the conductive member 1703 by merely providing a step on the suction surface side of the electrostatic chuck. However, in order to increase the Coulomb force and the like as much as possible, it is desirable to reduce the interval between the adsorption electrode and the sample. Therefore, as illustrated in FIG. 17, it is desirable that the adsorption electrode is built in the protruding portion on the electrostatic chuck surface side, and as a result, a height difference is provided between the adsorption electrode and the conductive member 1703. Further, since the two electrodes are separated in the height direction (Z direction), even if the two electrodes are brought close to each other in the XY direction, the risk of discharge does not increase, and the electric field correction is further improved. It is possible to perform electrode arrangement in consideration of the effect.

  Subsequently, a second embodiment will be described with reference to FIG. In the second embodiment, in addition to the first embodiment, a shield electrode 1002 (third electrode) is provided on the outermost outer periphery of the attracting electrode 1001 inside the electrostatic chuck. The shield electrode 1002 is connected to a retarding power source 1003 having the same voltage as the wafer. In the first embodiment, the effect of charging on the surface of the electrostatic chuck is shielded by making it smaller than the outer diameter of the wafer, but the electric field generated by the chucking electrode of the electrostatic chuck also has a slight influence on the vicinity of the wafer periphery. give. If a retarding voltage having the same voltage as that of the wafer is applied to the outermost electrode, the electric field 1004 generated by the adsorption electrode can be shielded. However, in such a configuration, the adsorption area of the electrostatic chuck is limited by the amount of the shield electrode. When it is desired to shield the influence of the internal electrode while ensuring a sufficient suction area, an insulating member 1104 is not provided between the electrostatic chuck 1101 and the conductive member 1102, as shown in FIG. A shield member 1103 may be provided. By adopting such a configuration and applying a retarding voltage 1105 to the shield member, it is possible to secure a sufficient adsorption area of the electrostatic chuck while shielding the influence of the electric field of the internal electrode. Although the shield member has been described as a separate member from ESC (Electrostatic Chuck) here, the shield member may be formed by applying a conductive coating to the outer periphery of the ESC.

  FIG. 12 shows a measurement flow of CD-SEM in the first embodiment and the second embodiment. First, a wafer is carried into the sample chamber, and an electrostatic chuck and a retarding voltage are applied (step 1201). Thereafter, a predetermined voltage Vc is applied to the conductive member (step 1202). The voltage value Vc to be applied at this time is an optimum voltage value determined by the gap d between the conductive member and the wafer side surface, the height h1 of the conductive member, and the height h2 of the stepped portion when there is a step. Vc (d, h1, (h2)) is selected. Thereafter, length measurement processing is performed at each measurement point (step 1203). When the length measurement processing is completed, the wafer is unloaded (step 1204). With such a flow, an appropriate voltage is always applied to the conductive member during wafer observation, so it is possible to ensure an observable range without being aware of the measurement position when measuring anywhere on the wafer. Is possible.

  Next, as a third embodiment, a measurement flow of a CD-SEM for further expanding the length measurable region in addition to the first embodiment and the second embodiment will be described. The electric field near the wafer edge tends to be affected by the disturbance of the electric field at the stepped portion as it approaches the edge. Therefore, the optimum voltage to be applied to the correction electrode is slightly different depending on a slight difference in distance from the wafer edge at the observation position. FIG. 13 is a measurement flow of CD-SEM in the third embodiment. Rather than applying the correction voltage Vc as a constant value after applying the retarding voltage, the distance r from the wafer center is calculated for each movement to each measurement point (beam irradiation point), and the optimum voltage Vc corresponding to the distance r is calculated. (D, h1, (h2), r) is applied to the conductive member (step 1301). By adopting such a flow, an optimum correction voltage can be applied for each observation position, so that the length measurement possible region can be further expanded. In addition, since voltage resetting for each measurement is performed in parallel with stage movement, the apparatus throughput is not reduced.

  Subsequently, a fourth embodiment will be described. The position of the wafer placed on the electrostatic chuck varies from the center of the electrostatic chuck by an error in the conveyance accuracy. This variation gives the same variation between the wafer edge and the gap between the conductive members installed outside the electrostatic chuck. When the conductive member and the wafer center are displaced, the gap between the wafer and the conductive member in the direction in which the wafer is displaced is relatively narrow, and the gap on the opposite side is relatively wide. The size of the gap varies depending on the location. A CD-SEM and a measurement flow that can correct such a gap variation due to the conveyance accuracy will be described with reference to FIGS. 14 and 15. After carrying the wafer into the sample chamber, first, a plurality of gaps between the wafer 1402 and the conductive member 1403 are imaged by a position detection device such as an optical microscope 1401 shown in FIG. 14 (step 1501), and the position of the wafer is obtained by image processing. Deviation amounts Δx and Δy are calculated (step 1502). When moving to each measurement point (Xn, Yn) (step 1503), not only the distance r from the wafer center but also the gap d between the wafer and the conductive member in the observation position Xn, Yn direction determined by the positional deviation amounts Δx, Δy. An optimum correction voltage Vc (h1, (h2), r, d (Δx, Δy, Xn, Yn)) corresponding to (Δx, Δy, Xn, Yn) is applied (step 1504). FIG. 16 shows the positional relationship of parameters for determining these optimum voltages Vc.

  The gap d (Δx, Δy, Xn, Yn) determined by the positional deviations Δx, Δy of the wafer 1601 and the observation positions Xn, Yn, the distance r from the center of the wafer, and the heights h1, (h2) of the conductive members 1602 The optimum voltage Vc is determined. By adopting such a flow, it is possible to perform correction in consideration of wafer conveyance variation, and it is possible to further expand the observable range. In this case, the wafer is displaced by an optical microscope, but may be performed by a line sensor.

  According to the configuration as described above, in the scanning electron microscope to which the electrostatic chuck is applied, without being affected by the internal electrode of the electrostatic chuck and the influence of charging on the electrostatic chuck accumulated during operation of the apparatus, It becomes possible to ensure a stable observation range.

1 Sample 2 Electrostatic Chuck 3 Sample Adsorption Surface 4 Adsorption Electrode 5 Electric Field Correction Power Supply 6 Electrode 7 Virtual Straight Line 8 First Surface 9 Second Surface

Claims (9)

In an electrostatic chuck mechanism having a sample adsorption surface for adsorbing a sample, and a first electrode to which a voltage for generating an adsorption force is applied between the sample adsorption surface and the sample.
A second electrode having a first surface below the first electrode; and a negative voltage application power source for applying a negative voltage to the second electrode, wherein the second electrode includes the first electrode An electrostatic chuck mechanism comprising a second surface surrounding the surface and serving as the uppermost surface of the second electrode positioned below the sample adsorption surface. In claim 1,
The electrostatic chuck characterized in that the second electrode has surfaces with different heights in a direction orthogonal to the sample adsorption surface, and the surfaces with the different heights become higher as the distance from the first electrode increases. mechanism. In claim 2,
The electrostatic chuck mechanism, wherein the second electrode is formed in a step shape or a slope shape. In claim 1,
An electrostatic chuck mechanism, wherein a third electrode is disposed between the first electrode and the second electrode. In claim 4,
The electrostatic chuck mechanism, wherein the same voltage is applied to the first electrode and the third electrode. In claim 1,
The electrostatic chuck mechanism, wherein the second electrode is disposed so as to surround the first electrode. In a charged particle beam apparatus comprising a beam column for irradiating a charged particle beam and a sample stage for moving a sample irradiated with the charged particle beam,
An electrostatic chuck mechanism comprising: a sample adsorption surface for adsorbing the sample; and a first electrode to which a voltage for generating an adsorption force is applied between the sample adsorption surface and the sample. The mechanism includes a second electrode having a first surface below the first electrode, and a negative voltage application power source that applies a negative voltage to the second electrode. A charged particle beam apparatus comprising: a second surface surrounding the first surface and serving as the uppermost surface of the second electrode located below the sample adsorption surface. In claim 7,
A charged particle beam apparatus comprising: a control device that adjusts a voltage to be applied to the second electrode in accordance with a location irradiated with the charged particle beam. In claim 8,
A charged particle beam comprising a position detection device for detecting the position of the sample, wherein the control device adjusts a voltage applied to the second electrode based on a position detection result of the position detection device. apparatus.

Images