JP6979107B2 - Charged particle beam device - Google Patents

Description

The present disclosure relates to a charged particle beam device, and more particularly to a charged particle beam device including an electrostatic chuck mechanism.

Some scanning electron microscopes for measuring and inspecting a semiconductor wafer employ an electrostatic chuck mechanism as a support mechanism for supporting the semiconductor wafer to be irradiated with an electron beam. The electrostatic chuck applies a voltage to a metal electrode provided inside to generate positive and negative charges on the object to be adsorbed such as a semiconductor wafer and the surface of the electrostatic chuck, and the Coulomb force acting during this period causes the object to be adsorbed. Fix it. In Patent Document 1, a ring-shaped correction electrode surrounding the outer peripheral portion of the sample is described in order to suppress the influence of the electric field generated in the vicinity of the peripheral portion of the object to be adsorbed on the electron beam by applying a voltage to the electrostatic chuck. The electrostatic chuck mechanism provided with the above is disclosed. An electric field that can be represented by equipotential lines parallel to the sample surface is formed between the electrostatic chuck mechanism and the objective lens directly above the sample, but the electric field changes at the end of the sample, and the electric field changes. It is a deflection electric field that deflects the beam. By applying an appropriate voltage to the correction electrode, the beam deflection can be suppressed.

Japanese Patent Application Laid-Open No. 2015-176683 (Corresponding US Patent USP9,401,297)

The sample adsorption surface of the electrostatic chuck is formed of, for example, an insulating ceramic as a dielectric layer. On the other hand, the correction electrode is formed of conductive aluminum or the like. In order to suppress the deflection action of the beam uniformly regardless of the irradiation position of the beam, it is necessary that the sample adsorption surface of the electrostatic chuck and the correction electrode are aligned with high accuracy. As described, the member forming the dielectric layer and the correction electrode are directly coupled. However, according to the study by the inventors, in such a configuration, when a sample having a temperature different from that of the dielectric layer is mounted on the adsorption surface, heat is applied to the correction electrode attached to the member forming the dielectric layer and the dielectric layer. Was transmitted, and it was found that warpage may occur like bimetal. When the suction surface that supports the sample is warped, the sample is also deformed along the suction surface. If the sample is warped, the height of the sample will change and the beam may not be in focus. Patent Document 1 does not discuss such suppression of the warp of the adsorption surface.

Below, we propose a charged particle beam device for the purpose of suppressing sample deformation caused by mounting samples with different temperatures on the adsorption surface of the electrostatic chuck mechanism.

As one aspect for achieving the above object, it is a charged particle beam device provided with an electrostatic chuck mechanism, and a sample irradiated with a charged particle beam is moved relative to an irradiation position of the charged particle beam. The stage, an insulator arranged on the stage and constituting the dielectric layer of the electrostatic chuck, a first support member for supporting the insulator on the stage, and the insulation surrounding the sample. We propose a charged particle beam device provided with a ring-shaped electrode that is non-contactly installed on the body and to which a predetermined voltage is applied, and a second support member that supports the ring-shaped electrode.

According to the above configuration, even when there is a temperature difference between the sample arranged on the electrostatic chuck and the suction surface of the electrostatic chuck, it is possible to effectively suppress the sample deformation.

The figure which shows the outline of the charged particle beam device (measurement SEM). The figure which shows the outline of the holding mechanism (electrostatic chuck mechanism). The figure which shows the outline of the optical system of the charged particle beam apparatus. The figure which shows the example which corrects the electric field at the end of a sample by applying a voltage to a correction electrode. The figure which shows the electrostatic chuck mechanism which attached the correction electrode to the ceramic which constitutes a dielectric layer. The figure explaining the principle of sample deformation which occurs when a high temperature sample is placed on an electrostatic chuck. The figure which shows an example of the electrostatic chuck mechanism which has the correction electrode support member (second support member) different from the support member (first support member) of an electrostatic chuck. The figure which shows an example of the electrostatic chuck mechanism which supports a correction electrode by a plurality of correction electrode support members. The figure which shows an example of the electrostatic chuck mechanism which has the flatness correction material which keeps a correction electrode flat. The figure which shows an example of the electrostatic chuck mechanism provided with the correction electrode made of the mixed material of aluminum and metallic silicon. The figure which shows an example of the electrostatic chuck mechanism provided with the correction electrode made from the mixture of Ni-plated aluminum and metallic silicon.

The embodiments described below relate to the electrostatic chuck mechanism and the semiconductor inspection / measurement device, and even when there is a temperature difference between the device and the observation target, the focus time on the observation target is increased and the throughput is reduced. The electrostatic chuck mechanism and its peripheral mechanism, which can suppress this, relate to the mechanism.

In the device manufacturing line, a device applying a scanning electron microscope is used for measuring the dimensions of fine patterns and inspecting defects on the device. For example, a length measurement SEM (Critical-Dimension Scanning Electron Microscope: CD-SEM) is used for dimension measurement of a gate or contact hole of a semiconductor device (hereinafter referred to as length measurement), and a defect inspection SEM or the like is used for defect inspection. .. In addition, scanning electron microscopes have come to be used for continuity inspection of deep holes for wiring by utilizing potential contrast.

FIG. 1 is a diagram showing an electron microscope which is one aspect of a charged particle beam device. In the following description, it is assumed that the electron microscope is the length measuring SEM. The length measuring SEM 100 includes a stage 200 having a holding mechanism 201 for holding the observation target 202 and having a function of driving the inside of the device in a multiaxial direction, an electron optical system 300 for emitting electrons to the observation target 202, and a vacuum chamber 101. I have. The dotted line in the figure is a notch line in which the vacuum chamber 101 is cut out in order to visualize the inside. The length measuring SEM 100 takes in the observation target 202 from the sample stocker 104 outside the apparatus along the movement path 105 of the observation target, holds the observation target 202 by using the electrostatic chuck 201 provided in the stage 200, and holds the observation target 202 using the electrostatic chuck 201 provided in the stage 200. After positioning with respect to the object, the electron optics system 300 is focused on a desired portion of the observation target 202 to measure the length.

Next, the holding mechanism of the observation target will be described. FIG. 2 shows an outline of the holding mechanism. In the length measurement SEM, the electrostatic chuck 203 may be used as a holding mechanism for a semiconductor wafer (hereinafter referred to as a wafer) to be observed. The electrostatic chuck 203 has a disk-like shape mainly formed of ceramics constituting the dielectric layer, and by applying a voltage to a metal electrode provided inside, positive and negative are positive and negative on the surface of the wafer and the electrostatic chuck 203. A charge is generated, and the wafer is fixed by the Coulomb force acting during this period.

Although the wafer is molded with extremely high flatness, the suction surface of the electrostatic chuck 203 is also molded with extremely high flatness (± several μm) so as not to cause deformation of the wafer. The suction surface of the electrostatic chuck 203 is approximately the same size as the wafer. The electrostatic chuck 203 is fixed to the stage 200 via the chuck support portion 402. Further, a correction electrode 204 is provided around the electrostatic chuck 203. The correction electrode 204 is fixed to the electrostatic chuck 203. Details of the correction electrode 204 will be described later.

The outline of the electro-optical system 300 is shown in FIG. The primary electron beam 322 (indicated by a broken line) emitted from the electron gun 301 due to the voltage of the extraction electrode 302 passes through the condenser lens 303, the scanning deflector 305, the aperture 306, the objective lens 309, etc., and is converged and deflected for observation. The inspection position of the target wafer 205 is irradiated. The condenser lens 303, the scanning deflector 305, the aperture 306, the objective lens 309, and the shield electrode 316 are formed in an axially symmetric shape with the optical axis 318 as the central axis.

A deceleration voltage (hereinafter referred to as a retarding voltage) is applied to the wafer 205 from the retarding power supply 326 for deceleration of the primary electron beam 322. A secondary electron beam 324 (indicated by a broken line) is generated from the wafer 205 by irradiation with the primary electron beam 322, and is accelerated by the retarding voltage applied to the wafer 205 and moves upward. The accelerated secondary electron beam 324 is deflected by the E-cross B deflector 308 and is incident on the secondary electron detector 314. In this secondary electron detector 314, the incident secondary electrons 324 are converted into an electric signal, amplified by a preamplifier (not shown) to become a brightness modulation input for the signal of the inspection image, and the SEM image which is the image data of the inspection area. Is obtained.

By the way, when the length measuring SEM inspects other than the end portion of the wafer 205, for example, the central portion, the equipotential surface in the vicinity of the wafer 205 is formed parallel and flat along the surface of the wafer 205. The distribution is axisymmetric with the axis 318 as the central axis. However, when inspecting the end portion of the wafer 205, the surface of the wafer 205 disappears at the end portion, and an equipotential surface cannot be formed along the surface of the wafer 205 outside the end portion. The axial symmetry of the equipotential surface may be disturbed. If the axial symmetry of the equipotential surface is disturbed, the primary electron beam 322 is bent, and the primary electron beam 322 is located at a position away from the position where the optical axis 318 and the surface of the wafer 205 intersect, which is the position to be originally inspected on the wafer 205. The electron beam 322 hits it, and as a result, the part to be observed shifts.

In order to suppress this irradiation position shift, a correction electrode 204 is installed around the wafer 205 of the stage 200, and a voltage-variable DC power supply 348 is connected to the correction electrode 204. The analysis unit 327 calculates the set voltage of the DC power supply 348 according to the distance 1000 between the optical system and the surface of the wafer 205, which is the distance between the optical system 300 and the surface of the wafer 205, and the irradiation conditions of the primary electron beam 322, and the control unit 329. (Control device) controls the DC power supply 348 to the set voltage.

FIG. 4 is a diagram showing an example in which the electric field at the edge of the sample is corrected based on the voltage applied to the correction electrode 204. By applying a voltage to the correction electrode 204, an electric field can be formed so that a flat equipotential surface parallel to the wafer surface is formed so that the wafer surface exists up to the space 349 outside the end of the wafer 205. can. By correcting the electric field that is non-axis target for the electron beam optical axis in this way, it is possible to appropriately irradiate the portion to be observed with the beam.

FIG. 5 is a diagram showing an example in which the correction electrode is directly attached to a ceramic electrostatic chuck (AA cross section of FIG. 2). In FIG. 5, the aluminum correction electrode 400 is directly fixed to the electrostatic chuck 203 with a screw 401 to prevent the relative distances from fluctuating from the design value. Since the equipotential surface is affected by the distance between the wafer and the aluminum correction electrode 400, it is desirable that these relative distances do not deviate from the design values. However, since the aluminum correction electrode 400 does not come into contact with the wafer, it cannot be directly positioned. Therefore, in the example of FIG. 5, the wafer and the aluminum correction electrode 400 are directly fixed to the electrostatic chuck 203 which is contact-fixed to the wafer. The relative distance to and from is not deviated from the design value. The reason for using the aluminum correction electrode 400 is to prevent the electrons emitted from the electron optical system from being adversely affected by using the magnetic material. Aluminum is also used for the stage for the same reason.

The aluminum correction electrode 400 or the electrostatic chuck 203 may be provided with a constant temperature device 403 so that the temperature does not change when there is a relative temperature difference with the wafer. However, according to the studies by the inventors, it is difficult to completely suppress the temperature change even if the constant temperature device 403 is provided, and the temperature of the aluminum correction electrode 400 and the electrostatic chuck 203 is increased by the heat conduction accompanying the contact of the wafer. It was found to change and transform into a mortar or dome-shaped warped shape.

FIG. 6 shows a cross-sectional view of the aluminum correction electrode 400 and the electrostatic chuck 203 when the wafer 205 having a high relative temperature is taken in. The mechanism of deformation is that the ceramic electrostatic chuck 203 having a difference in the coefficient of linear expansion and the aluminum correction electrode 400 are deformed in a bimetal manner due to the difference in the amount of elongation due to the temperature change. The coefficient of linear expansion is about 7 × 10 ⁻⁶ [/ K] ^{for ceramics and about 2 × 10 −5} [/ K] for aluminum. Since the wafer 205 attracted and fixed to the electrostatic chuck 203 is thinner and more easily deformed than the electrostatic chuck 203 and the aluminum correction electrode 400, the wafer 205 is deformed to follow the electrostatic chuck 203. Generally, heat propagation includes conduction, convection, and radiation, but there is no convection because the area around the wafer is vacuum, and radiation has almost no effect at around 20 ° C, which is the general usage environment for length measurement SEM. Since there is no such thing, conduction is the main cause of heat propagation between the wafer 205, the aluminum correction electrode 400, and the electrostatic chuck 203. A plurality of rectangular chips (not shown) are formed on the wafer 205 over almost the entire region. When there are a plurality of length measurement points, the length measurement SEM moves the stage 200 to position a new length measurement point of the wafer 205 with respect to the electro-optical system, then focuses the electro-optical system and repeats the length measurement. .. Therefore, as shown in FIG. 6, when the distance 1000 between the current measuring point and the optical system of the new measuring point increases or decreases, the time required for focusing increases, and as a result, the throughput of the length measuring SEM decreases.
Therefore, in this embodiment, a charged particle beam apparatus provided with an electrostatic chuck mechanism capable of suppressing sample deformation due to a temperature difference between the wafer and the chuck mechanism will be described. If the sample deformation can be suppressed, the reduction in throughput can be suppressed.

In this embodiment, the ceramic electrostatic chuck and the aluminum correction electrode are non-contact on a stage that can be driven in multiple axial directions (at least two directions) equipped with an electrostatic chuck that holds the wafer and an aluminum correction electrode. At the same time, a fixing member (support member) for fixing each independently was provided. If the electrostatic chuck and the aluminum correction electrode are not directly fixed, there is a possibility that the relative positions will shift between the two. It is preferable to suppress by controlling the voltage variable DC power supply applied to the correction electrode.

According to the above configuration, the inspection property of the outer peripheral portion of the wafer is good, and it is possible to suppress the reduction of the throughput even when the length of the wafer having a temperature difference is measured.

FIG. 7 is a diagram showing an example of the electrostatic chuck mechanism, in which the ceramic electrostatic chuck and the aluminum correction electrode (ring-shaped electrode) are supported by different support members, so that the ceramic and aluminum are not in contact with each other. It is a figure which shows the example. In the example of FIG. 7, the ceramic electrostatic chuck 203 is fixed to the stage 200 via the chuck support portion 402, and the aluminum correction electrode 400 is fixed to the stage 200 via the correction electrode support portion 405 and the insulator 406. Fixed to. The insulator 406 prevents the voltage of the variable voltage DC power supply 348 from being conducted to the aluminum stage 200. In this configuration, an assembly error from the stage 200 occurs in the electrostatic chuck 203 and the aluminum correction electrode 400, respectively, so that the relative distance between the electrostatic chuck 203 and the aluminum correction electrode 400 directly fixes them. As a result, there is a possibility that the potential correction of the aluminum correction electrode 400 is not appropriate for the wafer 205 fixed to the electrostatic chuck 203. However, in this embodiment, the voltage adjustment of the variable voltage DC power supply 348 compensates for the fluctuation of the potential caused by the fluctuation of the relative distance, so that there is no problem. In this case, the correction amount according to the deviation is registered in advance as a table in a predetermined storage medium at a plurality of points at different positions on the wafer end, and the address information stored in the recipe which is the operation program at the time of measurement is stored. Therefore, it is advisable to read out the applied voltage and apply the voltage.

When the aluminum correction electrode 400 of FIG. 6 has a ring shape (in other words, when the cross-sectional shape is the same as that of FIG. 6 over the entire circumference of the electrostatic chuck 203), the electrostatic chuck 203, the stage 200, and the aluminum are used. The space 407 surrounded by the correction electrode 400 is substantially sealed. In this case, when the vacuum chamber is exhausted, it may be in a state of being accumulated in the space 407 surrounded by the air, and the vacuum exhaust may be hindered. By using the correction electrode support portion 408, it is possible to secure the flow of gas between the lower space of the electrostatic chuck and the space inside the vacuum chamber other than the lower space, and it is possible to perform rapid vacuum exhaust.

As described above, even when the temperature of the wafer is different from the temperature of the electrostatic chuck, the sample deformation can be suppressed, and as a result, the reduction of the throughput of the measuring device can be suppressed.

Next, a second embodiment will be described. In the first embodiment, the configuration in which the aluminum correction electrode 400 is supported by a plurality of support members has been described, but in the second embodiment, the aluminum correction electrode 400 is formed of a ceramic ring-shaped plate. An example in which the plate-shaped body is supported by a plurality of support members after being supported will be described. As shown in FIG. 9, the point that the aluminum correction electrode 400 is supported by the ceramic flatness correction material 409 for keeping the flatness is different from that of the first embodiment. Hereinafter, the effect of Example 2 will be described. Due to the influence of stress remaining on the aluminum-based material itself, the desired high flatness (± several tens of μm) may not be obtained by itself even if molding such as cutting is performed. When the aluminum correction electrode 400 having poor flatness and wavy is used, the equipotential surface to be formed flat along the wafer surface becomes wavy near the wafer end, and the voltage-variable DC power supply 348 is used. It becomes impossible to cope with the increase and decrease of the voltage. However, the aluminum correction electrode 400 fixed to the ceramic flatness correction material 409 molded from ceramic that can be easily processed into high flatness can suppress the wavy shape. As described above, the equipotential surface is flattened with high accuracy by interposing a ceramic ring-shaped plate-like body capable of forming a flat surface with high accuracy between the correction electrode and the support member. It becomes possible to do.

Next, a third embodiment will be described. FIG. 10 shows an outline of the third embodiment. In Example 3, it was molded with a mixed material of aluminum and metallic silicon (linear expansion coefficient 8 × 10 ^-6 [/ K]) having a linear expansion coefficient close to that of an electrostatic chuck. Since the correction electrode 500 made of a mixed material of aluminum and metallic silicon has a small difference in linear expansion coefficient from that of ceramic, it is possible to suppress the occurrence of sample deformation due to the difference between the wafer temperature and the chuck temperature. Further, since the mixed material of aluminum and metallic silicon is a non-magnetic conductor, it does not affect the electron beam.

Next, a fourth embodiment will be described. FIG. 11 shows a cross-sectional view of the correction electrode and the electrostatic chuck showing the outline of the fourth embodiment. In the third embodiment, the correction electrode 500 made of a mixed material of aluminum and metallic silicon was used, but in the fourth embodiment, the surface of the correction electrode 500 made of a mixed material of aluminum and metallic silicon is covered with Ni electroless plating. ing. The following effects can be obtained by using the correction electrode 501 made of a mixed material of Ni-plated aluminum and metallic silicon. Hereinafter, the effects of Example 4 will be described. The mixture of aluminum and metallic silicon is a porous material with a large number of pores. Therefore, gas may be released, which may reduce the degree of vacuum. In Example 4, the release of gas can be suppressed by covering the surface with Ni electroless plating. Moreover, since the electroless plating of Ni is non-magnetic, it does not affect the electron beam.

It is also effective to use an alumina ceramic material for the correction electrode and cover the surface with Ni electroless plating. In this case, since the ceramic does not have conductivity, the electroless plating of Ni is responsible for the conductivity of the correction electrode. Further, since the Ni-plated ceramic correction electrode is non-magnetic, it does not affect the electron beam.

100 ... Length measurement SEM
101 ... Vacuum chamber 104 ... Sample stocker 105 ... Movement path 200 of observation target ... Stage 201 ... Holding mechanism 202 ... Observation target 203 ... Electrostatic chuck 204 ... Correction electrode 205 ... Wafer 300 ... Electron optics system
301 ... Electron gun 302 ... Extraction electrode 303 ... Condenser lens 305 ... Scanning deflector 306 ... Aperture 308 ... E-cross B deflector 309 ... Objective lens 314 ... Secondary electron detector 316 ... Shield electrode 318 ... Optical axis 322 ... Primary electron Line 324 ... Accelerated secondary electron beam
326 ... Returning power supply 327 ... Analysis department
329 ... Control unit
332 ... Secondary electron beam 348 ... Variable voltage DC power supply
349 ... Outer space 350 ... Isopotential surface 400 ... Aluminum correction electrode 401 ... Screw 402 ... Chuck support 403 ... Constant temperature device 405 ... Correction electrode support 406 ... Insulator 407 ... Surrounded space 408 ... Columnar correction electrode Support portion 409 ... Ceramic flatness correction material 500 ... Aluminum and metal silicon mixed material correction electrode 501 ... Aluminum and metal silicon mixed material correction electrode 1000 whose surface is covered with Ni electroless plating ... Optical Distance from the system

Claims (9)

In a charged particle beam device equipped with an electrostatic chuck mechanism,
A stage for relatively moving with respect to the sample charged particle beam is irradiated irradiation position of the charged particle beam,
An insulator arranged on the stage and constituting the dielectric layer of the electrostatic chuck and
Surrounding said sample around, and a ring-shaped electrode to which a predetermined voltage is applied,
The ring-shaped electrodes is Ri Do from the nonmagnetic conductor, the stage is a charged particle beam apparatus which is a mixed material of aluminum and silicon metal. In claim 1,
The ring-shaped electrode is a charged particle beam device characterized by being aluminum. In claim 1,
The ring-shaped electrode is a charged particle beam device characterized by being a mixed material of aluminum and metallic silicon. In claim 3,
Both the ring-shaped electrode and the stage, or the either, the charged particle beam device, wherein a metal plating is applied to the part or surface. In claim 4,
A charged particle beam device characterized in that the metal used for the metal plating is nickel. In claim 5,
The metal plating is a charged particle beam device characterized by being electroless nickel plating. In a charged particle beam device equipped with an electrostatic chuck mechanism,
A stage that moves the sample irradiated with the charged particle beam relative to the irradiation position of the charged particle beam, and
An insulator arranged on the stage and constituting the dielectric layer of the electrostatic chuck and
A ring-shaped electrode surrounding the sample and to which a predetermined voltage is applied is provided.
A charged particle beam device, wherein the ring-shaped electrode is made of a non-magnetic conductor, and the stage is a ceramic having a metal plating on a part or a surface thereof. In claim 7,
A charged particle beam device characterized in that the metal used for the metal plating is nickel. In claim 8,
The metal plating is a charged particle beam device characterized by being electroless nickel plating.

Images