JP2006244875A - Mapping projection type electron beam device and defect inspection system using the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To evaluate a sample by high throughput by reducing axial chromatic aberration and enlarging transmission rate of secondary electron. <P>SOLUTION: The electron beam emitted from an electron gun 1-1 is irradiated on a sample 7-1 through a primary electro-optical system, by the above, the electron emitted from the sample is detected by a detector 12-1 through a secondary electro-optical system. A Wien filter 8-1 composed of multipole lens for correcting axial chromatic aberration is arranged between a magnifying lens 10-1 of the second electro-optical system and a beam splitter 5-1 splitting the primary electron beam and the secondary electron beam. The axial chromatic aberration generated on an objective lens 14-1 composed of electromagnetic lens, having a magnetic gap at sample side is corrected by the Wien filter. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a mapping projection type electron beam apparatus and a defect inspection system using the same, and more particularly to a mapping projection type electron beam capable of performing a defect inspection of a sample such as a semiconductor wafer at a high throughput. The present invention relates to an electron beam apparatus having an optical system and a defect inspection system using the apparatus.

  In general, in an electron beam apparatus that obtains an image of a sample by irradiating a sample to which a retarding voltage is applied with a primary electron beam and detecting secondary electrons emitted from the irradiation point, axial chromatic aberration is reduced. In order to make it small, it is necessary to increase the electric field strength applied between the sample surface and the objective lens.

However, if the electric field intensity on the sample surface is increased in order to reduce the axial chromatic aberration, a discharge occurs between the sample and the objective lens. For this reason, the above electric field strength has to be set relatively small, and there is a problem that axial chromatic aberration is relatively large after all.
The present invention has been made to solve the above-described problems, and its purpose is to reduce axial chromatic aberration and increase the transmittance of secondary electrons in an electron beam apparatus. It is to enable evaluation of a sample with high throughput.

In order to achieve the above-described object, in the electron beam apparatus according to the present invention having a mapping projection type electron optical system for inspecting the surface of a sample,
An electron gun that emits an electron beam;
A primary electron optical system for directing and irradiating the emitted electron beam on a sample;
A detector for detecting electrons;
A secondary electron optical system that guides an electron beam emitted from the sample by irradiation of the electron beam and obtained information on the surface of the sample to a detector;
At least one of the primary electron optical system and the secondary electron optical system includes a multipole lens.

In the above-described electron beam apparatus according to the present invention, the multipole lens is disposed between the magnifying lens of the secondary electron optical system and the beam separating means for separating the primary electron beam and the secondary electron beam. preferable. At this time, the objective lens closest to the sample surface is an electromagnetic lens with a magnetic gap provided on the sample side, and axial chromatic aberration caused by the electromagnetic lens is corrected by a multipole lens arranged in the secondary electron optical system. It is preferable to be able to do so.
The multipole lens is preferably disposed between the reduction lens of the primary electron optical system and beam separating means for separating the primary electron beam and the secondary electron beam. At this time, the primary electron optical system includes an axially symmetric lens, and the axially symmetric lens can be set so that the off-axis aberration at the field edge is equal to or less than a predetermined value set in advance, and is configured by an electromagnetic lens. The bore diameter of the lens is preferably 50 times or more the maximum diameter of the field of view.
The primary electron optical system further includes means for converting the electron beam from the electron gun into a multi-electron beam, and the detector is a secondary electron beam emitted from an irradiation point on the sample of the multi-electron beam. It is preferable to include a detection unit that individually detects the multi-electron beams.
Furthermore, the multipole lens is preferably a four-stage quadrupole lens.

The present invention also provides a defect inspection system for inspecting a defect on the surface of a sample, comprising an electron beam apparatus having the above-described mapping projection type electron optical system, and the system is detected by a detector of the electron beam apparatus. An image acquisition means for generating an image of the sample surface based on information on the sample surface contained in the electrons, and a defect evaluation means for inspecting the presence of defects on the sample surface by comparing the acquired image with a reference image. I have.
In the defect inspection system according to the present invention described above, the system further includes a sample transport system for transporting a sample, a sample placement unit for placing the sample, an XY stage for moving the sample placement unit two-dimensionally, A main chamber that houses the sample placement unit and the XY stage and is maintained in a vacuum state, and is located between the main chamber and the sample transport system. When the sample is moved from the sample transport system to the main chamber, the main chamber It is preferable to provide a load lock chamber for maintaining the vacuum state. At this time, the sample transport system preferably includes an electrostatic chuck having a function of preventing adhesion of particles to the sample.
Further, the XY stage preferably includes an air bearing having a differential exhaust mechanism in at least one axial direction.

The electron beam apparatus having the projection type electron optical system according to the present invention is configured as described above and is reduced by a lens system having two or more stages. Therefore, the optical path length is kept short and 1/100 to 1/2000. Therefore, the space charge effect can be reduced.
Furthermore, since the axial chromatic aberration is corrected using a multipole lens, the aperture angle can be increased, and an electron beam with a large beam current can be obtained while keeping the beam diameter small, thus improving the throughput. be able to.
Further, as described above, since the throughput of the electron beam apparatus can be improved, the inspection throughput can be similarly improved in the defect inspection system provided with the electron beam apparatus.

  FIG. 1A is an explanatory diagram showing an electron optical system of an electron beam apparatus having a mapping projection type electron optical system according to the first embodiment of the present invention. As shown in FIG. 1A, in this electron beam apparatus, an electron beam emitted from an electron gun 1-1 is converted into a condenser lens 2-1, a molded lens 4-1, and an aperture plate 3-1 provided with a multi-aperture. The resulting electron beam is deflected so as to be perpendicular to the sample surface 7-1 by the E × B separator 5-1, and aligned with the sample surface 7-1 by the objective lens 6-1. By focusing, the sample surface is irradiated in a rectangular shape by each electron beam. On the other hand, secondary electrons emitted from the sample surface 7-1 by this irradiation form an enlarged image at the position 18-1 by the objective lens 6-1, and the enlargement lens 10 by the Wien filter 8-1 including a multipole lens. An enlarged image is formed at the focal point 13-1 which is a position before -1.

  At this time, the axial chromatic aberration at the focal point 13-1 can be eliminated by the objective lens 6-1 and the Wien filter 8-1 without changing the position of the focal point 13-1. That is, the objective lens 6-1 which is an axially symmetric lens generates positive axial chromatic aberration, and the Wien filter 8-1 generates negative axial chromatic aberration. The lens position and the voltage applied to the electrodes are adjusted. By doing so, the absolute values of positive and negative axial chromatic aberration can be made equal. Thereby, the positive axial chromatic aberration caused by the objective lens 6-1 can be canceled by the negative axial chromatic aberration caused by the Wien filter 8-1. It should be noted that the axial chromatic aberration at the focal point 13-1 may be adjusted to a negative value close to zero, thereby canceling the small positive axial chromatic aberration generated in the magnifying lenses 10-1 and 11-1. it can.

The magnified image formed at the focal point 13-1 further passes through the magnifying lens 10-1, and a magnified image is generated in front of the magnifying lens 11-1. An image is formed on the detection surface. The detector 12-1 generates an optical signal corresponding to the formed enlarged image, and propagates the optical signal to an 8 × 8 or 12 × 12 CMOS image sensor (not shown) via an optical fiber (not shown). To do. The CMOS image sensor converts an optical signal into an electrical signal, and the electrical signal is processed by an image data processing device (not shown).
In the optical system, axial chromatic aberration occupies most of the total aberration. In the electron beam apparatus shown in FIG. 1, the axial chromatic aberration of the enlarged image formed at the focal point 13-1 is almost zero. Alternatively, since it can be adjusted to be a slightly negative value, the aperture angle can be increased while the aberration is kept below a certain value, so that the secondary electron transmittance can be improved.

The electron beam apparatus shown in FIG. 1A is configured not only to eliminate or reduce axial chromatic aberration, but also to reduce off-axis aberrations. That is, as shown in FIG. 1A, the objective lens 6-1 includes a magnetic lens 14-1 having a magnetic gap on the sample 7-1 side and an axially symmetric lens 15-1, and reduces off-axis chromatic aberration. In order to achieve this, the Bohr diameter D1 of the magnetic lens 14-1 is set to 50 times or more the field diameter. Further, the off-axis aberration can also be reduced by setting the distance D2 between the main surface of the objective lens 6-1 and the sample surface 7-1 to be larger than 10 mm.
In the electron beam apparatus shown in FIG. 1A, two or more stages of alignment deflectors may be disposed between the molded lens 4-1 and the E × B separator 5-1, which are on the primary electron optical system side. .

  FIG. 1B shows a cross-sectional view of the Wien filter 8-1. In this Wien filter, twelve electrodes 8-1-1 to 8-1-12 are arranged around the optical axis 8-1-15 so that independent voltages can be applied from the power source. Reference numeral 8-1-16 is an insulating spacer for supplying a voltage independently to each electrode. A magnetic field that satisfies the Wien condition can be generated by passing a current through the coils 8-1-13 and 8-1-14. The electron beam emitted from the image point 18-1 of the secondary electron optical system forms an image at the central portion 19-1 of the Wien filter 8-1 by appropriately setting the magnetic field and the voltage applied to the electrodes. Further, the image is formed again at the position 13-1 exiting the Wien filter 8-1. In this way, by forming an image twice, the third-order aberration is small and the axial chromatic aberration becomes a negative value over the visual field. This point is described in D. Ionovicin et. Al, Rev. Sci. Instrum, Vol. 75, No. 11, Nov, 2004.

FIG. 2 shows an electron optical system of an electron beam apparatus having a mapping projection type secondary electron detection system according to the second embodiment of the present invention. In this electron beam apparatus, the electron beam emitted from the electron gun 21-1 is focused by the condenser lens 23-1, and irradiates the aperture plate 25-1 having a multi-aperture. The multi-beams separated and formed in this way form a first reduced image via the shaping lens 26-1 and the reduction lens 28-1, and further reduced by the objective lens 34-1 to be placed on the sample 35-1. A reduced image is formed. At this time, there are four stages of quadrupole lenses 30-1-1, 30-1-2, 31-1-1 and 31-1 between the formation position of the first reduced image and the objective lens 34-1. -2 is provided, and these quadrupole lenses can correct axial chromatic aberration caused by the objective lens 34.
The electron beam of secondary electrons emitted from the sample 35-1 forms a first magnified image in front of the magnifying lens 36-1, and is further magnified by the magnifying lens 36-1, and is then detected by the detector 38-1. An image about 100 times larger than the image on the sample 35-1 is formed on the detection surface. The detector 38-1 generates an optical signal corresponding to the formed magnified image, and propagates the optical signal to the PMT (not shown) through an optical fiber. The PMT converts an optical signal into an electrical signal, and the electrical signal is processed by an image data processing device (not shown).
In the electron beam apparatus shown in FIG. 2, the primary electron optical system can include a Wien filter including a multipole lens instead of the quadrupole lens.

  FIG. 3 shows an electron optical system of an electron beam apparatus according to the third embodiment of the present invention. In this electron beam apparatus, the electron beam emitted from the electron gun 41-1 forms a reduced image at the position 54-1 via the axially symmetric lens 40-1 which is an electromagnetic lens, and diverges from the position 54-1 The electron beam is focused at a position 43-1 by a Wien filter 42-1 that focuses twice, and the electron beam that diverges therefrom is converted into an electron beam parallel to the optical axis by an axisymmetric lens 44. In this embodiment, the single-stage axially symmetric lens 44-1 is used, but the axially symmetric lens may have two or more stages. An electron beam parallel to the optical axis passes through a deflector 47-1 and a beam separator 48-1 composed of an electromagnetic deflector, and a plurality of apertures on a sample 46-1 via a multi-aperture lens 45-1. Focused as an electron beam.

  Since the electron gun crossover size is about 50 μmφ, it is necessary to create a 25 nmφ crossover image to obtain a 50 nmφ electron beam, and the electron beam diameter φ must be reduced to about 1/2000. However, it is reduced to about 1/38 by the axisymmetric lens 40-1 at the front stage at the position 54-1, and further reduced to about 1/60 by the multi-aperture lens 45-1 at the last stage. A reduction ratio of 2000 is obtained, and a multi-electron beam of 50 nmφ including aberration can be obtained. Further, if the multi-aperture plate 53-1 is provided at the rear stage of the axially symmetric lens 44-1, the aperture angle is reduced, the axial chromatic aberration is reduced, and a multi-electron beam of 50 nmφ can be obtained with a reduction ratio of about 1/1800. . In this case, since the optical path length can be shortened, blur due to the space charge effect can be reduced.

As described above, the electron beam apparatus according to the third embodiment also includes the Wien filter 42-1 for correcting axial chromatic aberration between the two axially symmetric lenses 40-1 and 44-1. . The Wien filter 42-1 has the same configuration as that described with reference to FIG. 1B, and the aperture of the multi-aperture plate 53-1 can be increased by correcting axial chromatic aberration using the Wien filter. Therefore, a multi-electron beam having a large current value can be obtained. Instead of the Wien filter, four stages of quadrupole lenses may be provided.
The multi-aperture lens 45-1 is formed by stacking three metal plates, and through these metal plates, as shown in FIG. 4, m rows and n columns of openings are formed in a matrix. . A positive high voltage is applied to the central metal plate. In the opening pattern shown in FIG. 4, if the ratio d1 / d2 between the opening diameter d1 and the opening pitch (distance between the centers of two adjacent openings) d2 is set to 2/3 to 4/5, It has been obtained by simulation that it is not affected by the opening.

Here, the optical axis from the electron gun 41-1 to the deflector 47-1 and the optical axis of the multi-aperture lens 45-1 are intentionally deviated, and the deflector 47-1 causes the electromagnetic beam separator 48- The axis is adjusted to position 1. The offset amount between the two optical axes (the distance between the two optical axes) is expressed as follows: D3 is the distance between the deflection center of the deflector 47-1 and the deflection center of the electromagnetic beam separator 48-1.
Offset amount = D3 · tan ^-1 15 °
Set to satisfy the relationship. In this example, the amount of deflection is set to be deflected by 15 ° by the primary electron beam and 16 ° by the secondary electron by the beam separator 48-1 consisting of an electromagnetic deflector. Also good.

  The secondary electron deflecting direction by the beam separator 48-1 composed of an electromagnetic deflector is opposite to the primary electron deflecting direction. Therefore, the direction of the primary electron beam with respect to the right direction in FIG. And the secondary electrons are separated from the primary electron beam. Since the secondary electron beam forms a sample image in the vicinity of the deflection main surface 50-1 (FIG. 4) of the deflector 48-1, the deflection chromatic aberration to the secondary electrons is small. The secondary electron beam separated from the primary electron beam by the beam separator 48-1 is detected by forming an enlarged image on the area detector 51-1 via the magnifying lens 49-1. Then, an image of the sample surface is synthesized by an image data processing device (not shown) with reference to irradiation arrangement information on the sample 46-1 of the primary electron beam which is a multi-beam. If the magnification ratio is insufficient, the number of stages of the magnifying lens in the secondary optical system may be increased.

The electron beam apparatus shown in FIG. 3 will be described in more detail with reference to FIG. 4 showing optical paths in the apparatus. The electron beam diverging from the crossover 41-1-1 formed by the electron gun 41-1 is made narrower and has a larger divergence angle by the axisymmetric lens 40-1 having a short focal length, and is incident on the Wien filter 42-1. A crossover is formed at the central portion 43-1 of the Wien filter 42-1. Upon exiting the Wien filter 42-1, a crossover is formed at a position 56-1 and incident on an axially symmetric lens 44-1, which becomes a parallel beam parallel to the optical axis.
In the parallel beam, the aperture angle of the beam is controlled by the multi-aperture of the multi-aperture plate 53-1, and the axial chromatic aberration is kept small. Next, it is deflected by 15 ° by the electromagnetic deflector 47-1 and deflected by 15 ° in the reverse direction by the beam separator 48-1 comprising the electromagnetic deflector. This is done by supplying to these two deflectors a deflection signal having the same absolute value but opposite deflection direction. Thereby, the multi-electron beam is again perpendicular to the z-axis direction, that is, the surface of the sample 46-1. Next, the multi-electron beam is irradiated onto the sample 46-1 via the multi-aperture lens 45-1, but the deflection by the deflectors 47-1 and 48-1 is the same angle in the opposite direction. If no chromatic aberration occurs and an axis alignment signal is superimposed on the deflector 47-1 or 48-1, the axis alignment between the multi-aperture lens 45-1 and the multi-aperture of the multi-aperture plate 53-1 is maintained. .

  The secondary electrons emitted from the sample 46-1 are focused by the applied acceleration electric field, become a small beam bundle, and pass through the objective lens 45-1 with almost no loss. Then, it is deflected in the direction of the secondary optical system by a beam separator 48-1 which is an electromagnetic deflector, and enters the magnifying lens 49-1. In FIG. 4, the trajectory of the secondary electron beam corresponding to one lens at the center of the multi-aperture lens 45-1 among the plurality of secondary electron beams from the sample 46-1 is indicated by a dotted line. The upper secondary electron beam forms an image at a position 50-1 after passing through the beam separator 48-1. Then, an enlarged image is formed on the detection surface of the detector 51-1 by the magnifying lens 49-1. The detector 51-1 generates an optical signal corresponding to the formed magnified image, and propagates the optical signal to a PMT (not shown) such as 8 × 8 or 12 × 12 through an optical fiber. The PMT converts an optical signal into an electric signal, which is processed by an image data processing device (not shown).

The voltage applied to each electrode of the Wien filter 42-1 and the current flowing through the coil are the crossover position 54-1, the axial position of the Wien filter 42-1, the magnitude of axial chromatic aberration to be corrected, the beam energy, etc. In consideration of the above, it may be set so that axial chromatic aberration disappears with the aid of simulation.
Moreover, in the electron beam apparatus of the first to third embodiments, the quadrupole lens may be configured by a multipole lens other than the quadrupole. In particular, it is preferable to use a 4 to 12-pole lens, and when the number of stages is four, the beam trajectory can be point-symmetrical in the first half and the second half to prevent the occurrence of secondary aberrations. Is the best.

Here, in the electron beam apparatus having the mapping projection type electron optical system, an operation effect obtained by eliminating the axial chromatic aberration using a multi-stage multipole lens will be described.
In the mapping projection type electron beam apparatus, the blur δc due to the Coulomb effect can be expressed as follows.
σc = I · L / (α · V ^3/2 )
Where I: electron beam current
L: Optical path length
α: Opening angle
V: Electron beam energy

  FIG. 5 shows the result of simulating the relationship between the on-sample aberration (nm) and the aperture angle α. The lens used in this simulation is a tablet type using two stages of electrostatic lenses, and is a lens in which lateral chromatic aberration is eliminated. Further, the voltage applied to the lens was set so that the electric field strength on the sample surface was about 1.5 kV / mm. According to the simulation result shown in FIG. 5, when the longitudinal chromatic aberration is removed, the aperture angle α can be improved about three times. In an electron beam apparatus having a mapping projection type electron optical system, axial chromatic aberration is dominant among the aberrations on the sample surface. It is shown that the aperture angle α can be improved by about 3 times without removing the above aberration.

In the above equation, when the blur δc is constant, if the aperture angle α is tripled, the beam current I can be tripled, and the transmittance is the square of the aperture angle, which is 9 times. . Therefore, since the beam current is tripled and the transmittance is nine times, the throughput expressed by these multiplications, that is, the inspection speed can be improved by 27 times. Further, by using LaB ₆ as an electron gun and adopting a multi-beam system as in the electron beam apparatus of the second embodiment, the inspection speed can be further improved.
Significantly improving the inspection speed is extremely important for an electron beam apparatus having a mapping projection type electron optical system.

In the SEM (scanning electron microscope), since the scanning deflector is arranged behind the multipole lens, regardless of the deflection scanning of the deflector, in the multipole lens arranged in the preceding stage, An electron beam passes only on the optical axis. Therefore, it is easy to correct axial chromatic aberration with a multipole lens.
On the other hand, in an electron beam apparatus having a mapping projection type electron optical system, it is necessary to reduce the aberration not only on the optical axis but also at a position away from the optical axis, so a multipole lens is simply used. It has been considered that it is difficult to correct longitudinal chromatic aberration by itself. Further, it has not been recognized that most of the aberrations that occur in an electron beam apparatus having a mapping projection type electron optical system are axial chromatic aberrations.
The present inventor can correct axial chromatic aberration suitably by using a multipole lens even in an electron beam apparatus having a mapping projection type electron optical system by performing actual machine tests and simulations using various parameters. As a result, the throughput can be greatly improved as described above.

Next, the overall configuration of an inspection system for evaluating a semiconductor wafer that can be used by incorporating the electron beam apparatus according to the present invention will be described.
6 and 7 are an elevation view and a plan view showing main components of the inspection system 1. The inspection system 1 is disposed between a cassette holder 10 that holds a cassette containing a plurality of wafers, a mini-environment device 20, a main housing 30, and a mini-environment device 20 and the main housing 30. , A loader housing 40 that defines two loading chambers, a stage device 50 that is placed in the main housing 30 and moves a wafer W, which is a wafer, and the wafer from the cassette holder 10 into the main housing 30. A loader 60 to be loaded on the arranged stage device 50 and an electron optical system 70 attached to the main housing 30 are provided, and these are arranged in a positional relationship as shown in FIGS. The electron beam apparatus of the present invention described above is incorporated as the electron optical system 70.

The inspection system 1 also includes a precharge unit 81 disposed in the vacuum main housing 30, a potential application mechanism that applies a potential to the wafer, an electron beam calibration mechanism, and positioning of the wafer on the stage apparatus 50. And an optical microscope 871 that constitutes an alignment control device 87 (shown in FIG. 26). The inspection system 1 further comprises a control device 2 for controlling the operation of these elements.
Below, the structure of each of the main elements (subsystems) of the inspection system 1 will be described in detail.

Cassette holder 10
The cassette holder 10 is provided with a plurality of cassettes c (for example, closed cassettes such as SMIF and FOUP manufactured by Assist) in which a plurality of wafers (for example, 25 wafers) are stored in a state of being arranged in parallel in the vertical direction. In this embodiment, two cassettes) are held. As this cassette holder, when a cassette is transported by a robot or the like and automatically loaded into the cassette holder 10, a structure suitable for the cassette holder is used, and when loading manually, an open cassette structure suitable for the cassette holder is used. Each can be selected and installed arbitrarily. In this embodiment, the cassette holder 10 is of a type in which the cassette c is automatically loaded. The cassette holder 10 includes, for example, an elevating table 11 and an elevating mechanism 12 that moves the elevating tail 11 up and down. The cassette c can be automatically loaded on the lifting table in the state shown by the chain line in FIG. 7, and is automatically rotated to the state shown by the solid line in FIG. Directed to the rotation axis of the first transport unit. Further, the lifting table 11 is lowered to a state indicated by a chain line in FIG.

  In another embodiment, as shown in FIG. 8, a plurality of 300 mm wafers W are accommodated in a grooved pocket (not shown) fixed inside the box body 501, and are transported and stored. is there. The substrate transport box 24 is connected to a rectangular tube-shaped box body 501 and an automatic opening / closing device for a substrate transport entrance door, and a substrate transport entrance door 502 capable of opening and closing a side opening of the box body 501 by a machine, The lid 503 is located on the opposite side of the opening and covers the opening for attaching and detaching the filters and the fan motor, and the groove-type pocket 507 for holding the wafer W. In this embodiment, the wafer is loaded and unloaded by the robot-type transfer unit 61 of the loader 60.

  The wafer stored in the cassette c is performed after or during the process of processing the wafer in the semiconductor manufacturing process. Specifically, a wafer that has undergone a film formation process, CMP, ion implantation, a wafer with a wiring pattern formed on the surface, or a wafer that has not yet been formed with a wiring pattern is stored in a cassette c for inspection. Is done. A large number of wafers accommodated in the cassette c are arranged side by side in parallel in the vertical direction so that a wafer at an arbitrary position in the cassette can be held by a first transfer unit to be described later. In addition, the arm of the first transport unit can be moved up and down.

Mini-environment device 20
FIG. 9 is an elevational view of the mini-environment device 20 as seen from a direction different from that in FIG. As shown in FIG. 9 and FIG. 6 and FIG. 7, the mini-environment device 20 includes a housing 22 that defines a mini-environment space 21 that is controlled in atmosphere, and clean air in the mini-environment space 21. A gas circulation device 23 for controlling the atmosphere by circulating a gas such as, a discharge device 24 that collects and discharges a part of the air supplied in the mini-environment space 21, and the mini-environment space 21 And a pre-aligner 25 for coarse positioning of a wafer as a sample.

  The housing 22 has a top wall 221, a bottom wall 222, and a peripheral wall 223 that surrounds the four circumferences, and has a structure that blocks the mini-environment space 21 from the outside. In order to control the atmosphere of the mini-environment space 21, the gas circulation device 23 is attached to the top wall 221 downward in the mini-environment space 21, as shown in FIG. The gas supply unit 231 that cleans the air) and flows the clean air in a laminar flow downward through one or more gas outlets (not shown) and the bottom wall 222. And a recovery duct 232 that recovers air that has flowed down toward the bottom, and a conduit 233 that connects the recovery duct 232 and the gas supply unit 231 to return the recovered air to the gas supply unit 231.

The laminar flow of the clean air, that is, the downward flow, that is, the downflow, is mainly supplied to flow through a transfer surface by a first transfer unit, which will be described later, arranged in the mini-environment space 21. It prevents dust that may be generated by the unit from adhering to the wafer. An entrance / exit 225 is formed in a portion of the peripheral wall 223 of the housing 22 adjacent to the cassette holder 10.
The discharge device 24 includes a suction duct 241 disposed below the transfer unit at a position below the wafer transfer surface of the transfer unit, which will be described later, a blower 242 disposed outside the housing 22, a suction duct 241 and a blower. And a conduit 243 for connecting the terminal 242 to the terminal 242. The discharge device 24 sucks a gas containing dust that may flow around the transport unit and may be generated by the transport unit through the suction duct 241, and the outside of the housing 22 through the conduits 243 and 244 and the blower 242. To discharge.

The pre-aligner 25 disposed in the mini-environment space 21 is arranged on an orientation flat formed on the wafer (referred to as a flat portion formed on the outer periphery of a circular wafer, hereinafter referred to as an orientation flat) or on the outer peripheral edge of the wafer. One or more V-shaped notches or notches formed are detected optically or mechanically, and based on this, the rotational position about the axis O ₁ -O ₁ of the wafer is approximately ± Pre-position with accuracy of 1 degree. The pre-aligner 25 constitutes a part of a mechanism for determining the coordinates of the wafer, which is a wafer, and is responsible for rough positioning of the wafer.

Main housing 30
As shown in FIGS. 6 and 7, the main housing 30 that defines the working chamber 31 includes a housing main body 32, and the housing main body 32 is a vibration isolating device, that is, a vibration isolating device disposed on the base frame 36. 37 is supported by a housing support device 33 mounted on 37. The housing support device 33 includes a frame structure 331 assembled in a rectangular shape. The housing main body 32 is disposed and fixed on the frame structure 331, and is connected to the bottom wall 321 mounted on the frame structure, the top wall 322, the bottom wall 321 and the top wall 322, and surrounds the four circumferences. The working chamber 31 is isolated from the outside. In this embodiment, the housing 32 main body and the housing support device 33 are assembled in a rigid structure, and vibrations from the floor on which the base frame 36 is installed are transmitted to the rigid structure. Stopping at 37. Of the peripheral wall 323 of the housing 32, a peripheral wall adjacent to the loader housing 40 is formed with a wafer entrance / exit 325.
The working chamber 31 is maintained in a vacuum atmosphere by a general-purpose vacuum device (not shown). A control device 2 that controls the operation of the entire inspection system 1 is disposed under the base frame 36.

In the inspection system 1, various housings including the main housing 30 are evacuated, and the evacuation system for that purpose includes a vacuum pump, a vacuum valve, a vacuum gauge, a vacuum pipe, and the like. The optical system, detector unit, wafer chamber, load lock chamber, etc. are evacuated according to a predetermined sequence. In each part, the vacuum valve is controlled so as to achieve a necessary degree of vacuum. Then, the degree of vacuum is constantly monitored, and when an abnormality occurs, emergency control is performed between the chambers or between the chamber and the exhaust system using an isolation valve or the like by an interlock function to ensure the necessary degree of vacuum in each part. As the vacuum pump, a turbo molecular pump is used for main exhaust, and a roots type dry pump is used for roughing. The pressure at the inspection place (electron beam irradiation part) is 10 ⁻³ to 10 ⁻⁵ Pa, preferably 10 ⁻⁴ to 10 ⁻⁶ Pa, which is one digit lower than that.

Loader housing 40
FIG. 10 shows an elevation view of the loader housing 40 as seen from a different direction than FIG. As shown in FIGS. 10, 6, and 7, the loader housing 40 includes a housing main body 43 that defines a first loading chamber 41 and a second loading chamber 42. The housing main body 43 includes a bottom wall 431, a top wall 432, a peripheral wall 433 that surrounds the four circumferences, and a partition wall 434 that partitions the first loading chamber 41 and the second loading chamber 42. The loading chamber is isolated from the outside. The partition wall 434 has an opening, that is, an entrance / exit 435 for transferring the wafer W between the two loading chambers. Further, entrances and exits 436 and 437 are formed in a portion of the peripheral wall 433 adjacent to the mini-environment device 20 and the main housing 30. The housing main body 43 of the loader housing 40 is placed and supported on the frame structure 331 of the housing support device 33. Therefore, the floor vibration is not transmitted to the loader housing 40.

  Although the entrance / exit 436 of the loader housing 40 and the entrance / exit 226 of the housing 22 of the mini-environment device 20 are aligned, the communication between the mini-environment space 21 and the loading chamber 41 is selected between the entrances / exits 436 and 226. A shutter device 27 is provided to prevent this. Further, the entrance / exit 437 of the loader housing 40 and the entrance / exit 325 of the housing main body 32 of the main housing 30 are aligned, but the communication between the loading chamber 42 and the working chamber 31 is selected between the entrances 436, 325. A shutter device 45 is provided to prevent the sealing. Further, the opening formed in the partition wall 434 is provided with a shutter device 46 that closes the opening by a door 461 and selectively blocks communication between the first and second loading chambers. These shutter devices 27, 45 and 46 are adapted to hermetically seal each chamber when in the closed state.

In the first loading chamber 41, a wafer rack 47 is disposed that horizontally supports a plurality of (two in this embodiment) wafers W separated vertically.
The first and second loading chambers 41 and 42 are controlled in atmosphere to a high vacuum state (the degree of vacuum is 10 ⁻⁵ to 10 ⁻⁶ Pa) by a general-purpose vacuum exhaust device (not shown) including a vacuum pump. Is done. In this case, the contamination of the wafer can be effectively prevented by maintaining the first loading chamber 41 as a low vacuum chamber in a low vacuum atmosphere and maintaining the second loading chamber 42 as a high vacuum chamber in a high vacuum atmosphere. it can. By adopting such a loading housing structure including two loading chambers, the wafer W can be transferred from the loading chamber into the working chamber without delay. By adopting such a loading chamber structure, the throughput of inspection for defects and the like is improved, and the degree of vacuum around the electron source, which is required to be kept in a high vacuum state, is set as high as possible. Can be in a state.

A vacuum exhaust pipe and a vent pipe (not shown) for an inert gas (for example, dry pure nitrogen) are connected to the first and second loading chambers 41 and 42, respectively. This achieves an inert gas vent (injecting the inert gas to prevent oxygen gas other than the inert gas from adhering to the surface) in the atmospheric pressure state in each loading chamber.
In the main housing 30 that uses an electron beam, a typical lanthanum hexaboride (LaB ₆ ) used as an electron source of the electron optical system 70, that is, an electron gun, is in a high-temperature state to such an extent that thermal electrons are emitted once. When heated to a high temperature, it is important not to make it contact with oxygen or the like as much as possible in order not to shorten its life. Since the atmosphere control as described above is performed before the wafer W is loaded into the working chamber in which the electron optical system 70 of the main housing 30 is disposed, the possibility of contact with oxygen is reduced. The possibility of shortening the lifespan is reduced.

Stage device 50
The stage apparatus 50 includes a fixed table 51 disposed on the bottom wall 321 of the main housing 30, a Y table 52 that moves in the Y direction (direction perpendicular to the paper surface in FIG. 6) on the fixed table, and a Y table. An X table 53 that moves in the X direction (left-right direction in FIG. 6), a rotary table 54 that can rotate on the X table, and a holder 55 that is arranged on the rotary table 54 are provided. The wafer W is releasably held on the wafer placement surface 551 of the holder 55. The holder 55 may have a general-purpose structure capable of releasably gripping the wafer W mechanically or by an electrostatic chuck method. The stage device 50 operates the plurality of tables 52 to 54 using a servo motor, an encoder, and various sensors (not shown), thereby causing the wafer W held by the holder 55 on the mounting surface 551. In the X direction, Y direction, and Z direction (vertical direction in FIG. 6) with respect to the electron beam irradiated from the electron optical system 70, and further in the direction around the axis perpendicular to the wafer support surface (θ direction). Can be positioned with high accuracy.

Note that the positioning in the Z direction may be performed by finely adjusting the position of the mounting surface on the holder 55 in the Z direction, for example. In this case, the reference position of the mounting surface is detected by a position measuring device (laser interference distance measuring device using the principle of an interferometer) using a fine-diameter laser, and the position is controlled by a feedback circuit (not shown). In addition to or instead of this, the position of the notch or orientation flat of the wafer is measured to detect the planar position and the rotational position of the wafer with respect to the electron beam, and the rotation table 54 is rotated by a stepping motor or the like capable of controlling a minute angle. . The wafer W may be directly placed on the rotary table 54 without providing the holder 55. Servo motors 521, 531 and encoders 522, 532 for the stage device 50 are disposed outside the main housing 30 in order to prevent dust generation in the working chamber 31 as much as possible.
Signals can be standardized by inputting in advance a rotation position or XY coordinate position of the wafer W with respect to the electron beam into a signal detection system or an image processing system described later.

Further, the wafer chucking mechanism provided in the holder is adapted to apply a voltage for chucking the wafer to the electrode of the electrostatic chuck, and has three points on the outer periphery of the wafer (preferably in the circumferential direction, etc.). It is designed to be positioned by pressing down. The wafer chuck mechanism includes two fixed positioning pins and one pressing crank pin. The clamp pin can realize automatic chucking and automatic release, and constitutes a conduction point for voltage application.
In this embodiment, the table that moves in the left-right direction in FIG. 7 is the X table and the table that moves in the up-down direction in FIG. 13 is the Y table, but the table that moves in the left-right direction in FIG. The table that moves in the vertical direction may be the X table.

Wafer Chucking Mechanism 1) Basic Structure of Electrostatic Chuck In order to focus the electron optical system on the sample surface accurately and in a short time, it is preferable to make the unevenness of the sample surface, that is, the wafer surface as small as possible. Therefore, the wafer is attracted to the surface of the electrostatic chuck manufactured with good flatness (preferably flatness of 5 μm or less).
The electrode structure of the electrostatic chuck includes a monopolar type and a bipolar type. The monopolar type is a method in which the wafer is attracted by conducting the wafer in advance and applying a high voltage (generally about several tens to several hundreds volts) between one electrostatic chuck electrode and the bipolar type. It is not necessary to conduct the wafer, and the wafer can be attracted only by applying positive and negative voltages to the two electrostatic chuck electrodes. However, in general, in order to obtain stable adsorption conditions, it is necessary to form a shape in which two electrodes are combined in a comb tooth shape, and the electrode shape becomes complicated.

  On the other hand, for specimen inspection, a predetermined voltage (retarding voltage) is applied to the wafer in order to obtain imaging conditions for the electron optical system or to make the specimen surface easy to observe with electrons. There is a need to. In order to apply this retarding voltage to the wafer and stabilize the potential on the wafer surface, the electrostatic chuck needs to be of the above-mentioned monopolar type. (However, it is necessary to make the electrostatic chuck function as a bipolar type until it is connected to the wafer with a conductive needle, as will be described later. Therefore, the electrostatic chuck has a switchable structure between a monopolar type and a bipolar type. If the potential on the surface of the wafer is not stable at a predetermined value in each inspection mode, the imaging conditions are not satisfied and a clear image cannot be obtained. Therefore, it is necessary to surely confirm the continuity of the wafer before applying the retarding voltage.

Therefore, the wafer must be brought into mechanical contact with the wafer for electrical conduction. However, there is an increasing demand for preventing contamination of the wafer, and it is required to avoid mechanical contact with the wafer as much as possible, and contact with the edge of the wafer may not be permitted. In such a case, conduction must be established on the back surface of the wafer.
Usually, a silicon oxide film is formed on the back surface of the wafer, and conduction cannot be obtained as it is. Therefore, by bringing two or more needles into contact with the back surface of the wafer and applying a voltage between the needles, the oxide film can be locally broken and conductive with the silicon of the wafer base material. The voltage applied to the needle is a DC voltage or an AC voltage of about several hundred volts. Further, the needle material is required to be non-magnetic, wear-resistant and a high melting point material, and tungsten or the like can be considered. It is also effective to coat the surface with TiN or diamond in order to provide further durability or prevent contamination of the wafer. It is also effective to apply a voltage between the needles and measure the current with an ammeter in order to confirm that conduction with the wafer has been achieved. By applying a retarding voltage after confirming continuity, the wafer surface can be brought to a desired potential, and an inspection satisfying the imaging conditions can be performed.

A chucking mechanism as shown in FIG. 11 is made from the above background. The electrostatic chuck includes electrodes 19. 1, 19. 2, which are preferably arranged in a comb-like shape for stably adsorbing the wafer W, and three pusher pins 19. 3 for delivering the wafer. Two or more conductive needles 19 and 4 for applying a wafer are provided. Further, a correction ring 19.5 and a wafer dropping mechanism 19.6 are disposed around the electrostatic chuck.
The pusher pins 19.3 protrude in advance from the surface of the electrostatic chuck when the wafer W as a sample is transferred by the robot hand. When the wafer W is placed on the wafer by the operation of the robot hand, And the wafer W is placed on the electrostatic chuck. When the wafer is taken out from the electrostatic chuck, the wafer W is transferred to the robot hand by performing the reverse operation. For the pusher pins 19.3, the surface material must be selected so that the wafer is not displaced or contaminated, such as silicon rubber, fluorine rubber, ceramics such as SiC or alumina, resin such as Teflon or polyimide, etc. It is desirable to use

  There are several methods for driving the pusher pins 19.3. One is a method of installing a non-magnetic actuator below the electrostatic chuck. This includes a method in which the pusher pin is directly linearly driven by an ultrasonic linear motor, and a method in which the pusher pin is linearly driven by a combination of a rotary ultrasonic motor and a ball screw or rack and pinion gear. In this method, the pusher mechanism is compactly integrated on the table of the XY stage on which the electrostatic chuck is mounted, but the wiring of the actuators, limit sensors, and the like becomes very large. These wires are connected from the XY-operating table to the sample chamber (main chamber or main housing) wall surface, but bend as the stage moves, so it is necessary to arrange them with a large bend R and take up space. End up. Moreover, since it becomes a particle generation source and the wiring needs to be replaced periodically, the number of use should be minimized.

Therefore, as another method, there is a method of supplying driving force from the outside. When the stage moves to a position where the wafer W is attached or detached, the shaft protruding into the vacuum via the bellows is driven by an air cylinder provided outside the chamber, and the shaft of the pusher drive mechanism provided below the electrostatic chuck It is supposed to press. The shaft is connected to the rack and pinion or the link mechanism inside the pusher drive mechanism, and the reciprocating movement of the shaft is interlocked with the vertical movement of the pusher pin. When the wafer W is transferred to and from the robot hand, the pusher pins 19.3 are raised by adjusting the speed to an appropriate level with a controller and pushing the shaft into a vacuum with an air cylinder.
The drive source of the shaft from the outside is not limited to the air cylinder, but may be a combination of a servo motor, a rack and pinion, and a ball screw. It is also possible to use an external drive source as the rotation axis. In that case, the rotary shaft is provided with a vacuum seal mechanism such as a magnetic fluid seal, and the pusher drive mechanism has a built-in mechanism for converting the rotation into a linear motion of the pusher.

The correction ring 19.5 has an action of maintaining the electric field distribution at the wafer end portion uniformly, and basically applies the same potential as that of the wafer. However, a potential slightly different from the wafer end potential may be applied in order to cancel out the influence of the minute gap between the wafer and the compensation ring and the minute difference between the wafer and the compensation ring surface height. The correction ring has a width of about 10 to 30 mm in the radial direction of the wafer, and can use a nonmagnetic and conductive material such as titanium, phosphor bronze, TiN or TiC coated aluminum.
The conductive needles 19 and 4 are supported by springs 19 and 7, and when the wafer is mounted on the electrostatic chuck, it is lightly pressed against the back surface of the wafer by the spring force. In this state, electrical continuity with the wafer W is obtained by applying a voltage as described above.

The electrostatic chuck main body is made up of non-magnetic planar electrodes 19, 19, 2, such as tungsten, and a dielectric formed thereon. As the dielectric material, alumina, aluminum nitride, polyimide, or the like can be used. In general, ceramics such as alumina are a perfect insulator having a volume resistivity of about 10 ¹⁴ Ωcm, so that no charge transfer occurs inside the material and a Coulomb force acts as an adsorption force. On the other hand, the volume resistivity can be reduced to about 10 ¹⁰ Ωcm by slightly adjusting the ceramic composition, and this causes the movement of electric charge inside the material, so that the wafer adsorption force is stronger than the Coulomb force. A so-called Johnson-Rahbek force acts. If the attractive force is strong, the applied voltage can be lowered accordingly, a margin for dielectric breakdown can be increased, and a stable attractive force can be easily obtained. In addition, by processing the electrostatic chuck surface into, for example, a dimple shape, even if particles adhere to the electrostatic chuck surface, the particles may fall into the valley portion of the dimple, which affects the flatness of the wafer. The effect of reducing the possibility can be expected.
From the above, the electrostatic chuck material is made of aluminum nitride or alumina ceramic whose volume resistivity is adjusted to about 10 ¹⁰ Ωcm, and the surface is formed as a set of convex surfaces by forming irregularities such as dimples on the surface. What is processed into about 5 μm is practical.

2) Chucking mechanism for 200/300 bridge tool There are cases where the apparatus is required to inspect two types of wafers of 200 mm and 300 mm without mechanical modification. In this case, the electrostatic chuck must chuck two types of wafers and place a correction ring in accordance with the wafer size on the peripheral edge of the wafer. FIGS. 11A, 11B, and 12 show structures for this purpose.
FIG. 11A shows a state where a 300 mm wafer W is mounted on the electrostatic chuck. A correction ring 19.1 having an inner diameter slightly larger than the size of the wafer W (a gap of about 0.5 mm) is positioned and mounted on a metallic ring-shaped part on the outer periphery of the electrostatic chuck by an inlay. The correction ring 19. 1 is provided with three wafer dropping mechanisms 19.2. The wafer dropping mechanism 19.2 is driven by a vertical drive mechanism that is linked to the drive mechanism of the pusher pins 19.3, and is supported so as to be rotatable around a rotation axis provided in the correction ring 19.1.

When the wafer W is received from the robot hand, the pusher pin drive mechanism operates to push up the pusher pins 19.3. At the same time, the wafer dropping mechanism 19.2 provided on the correction ring 19.1 is also rotated by receiving a driving force as shown in FIG. Then, the wafer dropping mechanism 19.2 forms a tapered surface for guiding the wafer W to the electrostatic chuck center. Next, after the wafer W is placed on the pushed-up pusher pins 19.3, the pusher pins 19.3 are lowered. By appropriately adjusting the timing of the driving force applied to the wafer dropping mechanism 19.2 with the lowering of the pusher pins 19.3, the position of the wafer W is corrected by the taper surface of the dropping mechanism 19.2 and the electrostatic chuck. The center of the wafer W and the center of the electrostatic chuck are placed on the top.
It is desirable to form a low friction material such as Teflon, preferably a conductive low friction material (for example, conductive Teflon, conductive diamond-like carbon, TIN coating) on the tapered surface of the dropping mechanism 19 or 2. Reference numerals A, B, C, D, and E in the figure are terminals (described later) for applying a voltage, and 19.4 detects that the wafer W is placed on the electrostatic chuck. The wafer conduction needle is pushed up by a spring 19.5.

FIG. 12 shows a state where a 200 mm wafer W is mounted on the same electrostatic chuck. Since the surface of the electrostatic chuck is exposed because the wafer diameter is smaller than that of the electrostatic chuck, the correction ring 20. 1 having a size that completely hides the electrostatic chuck is mounted. The positioning of the correction ring 20.1 is the same as in the case of the 300 mm correction ring.
A step is provided on the inner peripheral portion of the correction ring 20. 1 so as to be accommodated in the ring-shaped groove 20. This is a structure for concealing the surface of the electrostatic chuck with a conductor (correction ring 20. 1) so that the surface of the electrostatic chuck cannot be seen from the gap between the inner periphery of the correction ring 20. It is. If the electrostatic chuck surface is visible, when the electron beam is irradiated, the electric charge is charged on the surface of the electrostatic chuck, and the potential of the sample surface is disturbed.

To replace the correction ring 20. 1, a correction ring replacement place is provided at a predetermined position in the vacuum chamber, and a correction ring having a required size is transported by the robot and attached to the electrostatic chuck (on the inlay portion). Insert).
The 200 mm correction ring is also provided with a wafer dropping mechanism 20. 2 as in the case of 300 mm. A relief is formed on the electrostatic chuck side so as not to interfere with the wafer dropping mechanism 20. The wafer mounting method on the electrostatic chuck is exactly the same as in the case of 300 mm. Reference numerals A, B, C, D, and E are terminals for applying a voltage, 20 and 3 are push pins similar to the push pins 19 and 3, and 20 and 4 are similar to the wafer conduction needles 19.4. Wafer conduction needle.

  FIGS. 13A and 13B are diagrams schematically showing the configuration of an electrostatic chuck that can handle both 300 mm wafers and 200 mm wafers, and FIG. ) Shows a state in which a 200 mm wafer is placed. As understood from FIG. 13A, the electrostatic chuck has a size capable of mounting a 300 mm wafer, and as shown in FIG. 13B, the central portion of the electrostatic chuck is 200 mm. Grooves 20 and 6 into which the inner peripheral portion of the correction ring 20 and 1 fits are provided so as to surround the wafer. Reference signs A, B, C, D, and E are terminals for applying a voltage.

  In the case of the electrostatic chuck shown in FIGS. 13A and 13B, whether the wafer is placed on the electrostatic chuck, whether the wafer is correctly placed on the electrostatic chuck, and whether there is a correction ring. Etc. are detected optically. For example, an optical sensor is installed above the electrostatic chuck, and the wafer is placed horizontally by detecting the optical path length when the light emitted from the optical sensor is reflected by the wafer and returns to the optical sensor again. It can be detected whether it is mounted or tilted. The presence or absence of the correction ring is determined by providing an optical transmitter that obliquely illuminates an appropriate point in the place where the correction ring should be placed and an optical receiver that receives the reflected light from the correction ring. Can be detected. Further, a combination of an optical transmitter for obliquely irradiating an appropriate point where a correction ring for a 200 mm wafer is placed, an optical receiver for receiving reflected light from the correction ring, and a correction ring for a 300 mm wafer A combination of an optical transmitter that obliquely illuminates an appropriate point on the place where the light is placed and an optical receiver that receives reflected light from the correction ring, and which optical receiver receives the reflected light It is possible to detect which one of the 200 mm wafer correction ring and the 300 mm wafer correction ring is placed on the electrostatic chuck.

3) Wafer chucking procedure The wafer chucking mechanism having the above-described structure chucks a wafer according to the following procedure.
(1) A correction ring suitable for the wafer size is transported by a robot and mounted on an electrostatic chuck.
(2) The wafer is placed on the electrostatic chuck by the wafer transfer by the robot hand and the vertical movement of the pusher pin.
(3) An electrostatic chuck is applied in a bipolar manner (positive and negative voltages are applied to the terminals C and D) to attract the wafer.
(4) A predetermined voltage is applied to the conduction needle to destroy the insulating film (oxide film) on the back surface of the wafer.
(5) The current between the terminals A and B is measured, and it is confirmed whether or not electrical connection with the wafer is obtained.
(6) The electrostatic chuck is shifted to monopolar adsorption. (Apply the same voltage to terminals A and B as GRD and terminals C and D)
(7) While maintaining the potential difference between the terminal A (or B) and the terminal C (or D), the voltage at the terminal A (or B) is lowered and a predetermined retarding voltage is applied to the wafer.

Device Configuration for 200/300 Bridge Tool FIGS. 14 and 15 show a configuration for making an apparatus capable of inspecting both 200 mm wafers and 300 mm wafers without mechanical modification. Hereinafter, differences from the 200 mm wafer or 300 mm wafer dedicated apparatus will be described.
Wafer cassettes according to the wafer size and the type of wafer cassette determined by the user specifications can be installed at the wafer cassette installation location 21.1 which is exchanged for each specification such as 200/300 mm wafer, FOUP, SMIF, and open cassette. It has become. The atmospheric transfer robots 21 and 2 are provided with a hand that can cope with different wafer sizes, that is, a plurality of dropping portions of the wafer are provided in accordance with the wafer size, and are mounted on the hand at a position corresponding to the wafer size. It is like that. The atmospheric transfer robot 21. 2 sends the wafer from the installation location 21. 1 to the pre-aligner 21. 3 to adjust the orientation of the wafer, and then removes the wafer from the pre-aligner 21. 3 and enters the load lock chamber 21. send.

  The wafer racks inside the load lock chambers 21 and 4 have the same structure, and a plurality of drop portions corresponding to the wafer size are formed on the wafer support portion of the wafer rack. The height of the robot hand is adjusted so that the wafer loaded on the drop-in part that fits the size of the wafer is inserted into the wafer rack, and then the robot hand is lowered to support the wafer. The wafer is placed on a predetermined drop portion of the part.

  The wafers placed on the wafer racks in the load lock chambers 21 and 4 are then taken out from the load lock chambers 21 and 3 by the vacuum transfer robots 21 and 6 installed in the transfer chambers 21 and 5 to be sample chambers 21. -It is conveyed on stage 21 * 8 in 7. Similarly to the atmospheric transfer robots 21 and 2, the hands of the vacuum transfer robots 21 and 6 also have a plurality of drop portions that match the wafer size. The wafer mounted on the predetermined drop portion of the robot hand is placed on the electrostatic chuck on which the correction rings 21 and 9 suitable for the wafer size are mounted in advance on the stages 21 and 8, and is attracted and fixed by the electrostatic chuck. Is done. The correction rings 21 and 9 are placed on correction ring racks 21 and 10 provided in the transfer chambers 21 and 5. Accordingly, the vacuum transfer robots 21 and 6 take out the correction rings 21 and 9 suitable for the wafer size from the correction ring racks 21 and 10 and transfer them onto the electrostatic chuck, and positioning inlay portions formed on the outer periphery of the electrostatic chuck. After the correction rings 21 and 9 are fitted into the wafer, the wafer is placed on the electrostatic chuck.

When replacing the correction ring, the reverse operation is performed. That is, the correction rings 21 and 9 are removed from the electrostatic chuck by the robots 21 and 6, the correction rings are returned to the correction ring racks 21 and 10 in the transfer chambers 21 and 5, and the correction ring suitable for the wafer size to be inspected is corrected. Transport from the ring racks 21 and 10 to the electrostatic chuck.
In the inspection apparatus shown in FIG. 14, since the pre-aligner 21 · 3 is disposed near the load lock chambers 22 · 4, the wafer is not sufficiently aligned and the correction ring cannot be mounted in the load lock chamber. However, it is easy to return the wafer to the pre-aligner for realignment, and there is an advantage that time loss in the process can be reduced.

FIG. 15 is an example in which the location of the correction ring is changed, and the correction ring racks 21 and 10 are omitted. In the load lock chamber 22.1, wafer racks and correction ring racks are formed in a hierarchy, and these can be installed in an elevator and moved up and down. First, in order to install a correction ring suitable for the wafer size to be inspected in the electrostatic chuck, the elevator of the load lock chamber 22.1 is moved to a position where the vacuum transfer robots 21 and 6 can take out the correction ring. Then, when the correction ring is installed on the electrostatic chuck by the vacuum transfer robots 21 and 6, the elevator is operated so that the wafer to be inspected can be transferred, and the wafers are taken out from the wafer rack by the vacuum transfer robots 21 and 6. After that, it is placed on an electrostatic chuck. In the case of this configuration, an elevator is required for the load lock chamber 22.1, but the vacuum transfer chamber 21.5 can be formed small, which is effective in reducing the footprint of the apparatus.
Note that the sensor for detecting whether or not a wafer is present on the electrostatic chuck is preferably installed at a position that can accommodate both different wafer sizes. A plurality of sensors that function may be arranged for each wafer size.
Comparison inspection and alignment are performed using the algorithm as described above.

Here, the overall procedure for defect inspection will be described.
As shown in FIG. 16, the defect inspection is performed by setting the conditions of the electron optical system (setting the imaging magnification, etc.), moving the stage while irradiating the electron beam, and performing TDI scan imaging (FIG. 17). In accordance with the inspection conditions (array inspection conditions, random inspection conditions, inspection area), defect inspection is performed in real time by an inspection dedicated processing unit (IPE).
In the inspection recipe, the conditions of the electron optical system, the inspection target die, the inspection area, the inspection method (random / array), and the like are set (FIGS. 18A and 18B).
In addition, in order to acquire a stable image for defect inspection, EO correction that suppresses blurring of a captured image due to positional deviation or speed unevenness, and an error between an ideal die map arrangement and an actual die position are absorbed. Die position correction and focus adjustment that complements the focus value of the entire area of the wafer using focus values measured in advance at finite measurement points are simultaneously performed in real time.
In the scanning operation for defect inspection, in addition to inspecting the entire area of the inspection target die (FIG. 19), as shown in FIG. 20, thinning inspection can be performed by adjusting the amount of step movement in the direction perpendicular to the scanning direction. (Reduced inspection time).

After the inspection is completed, the number of defects, the position of the die including the defect, the defect size, the defect position in each die, the defect type, the defect image, and the comparison image are displayed on the display as the inspection result, and the information, recipe information, etc. It is possible to confirm and reproduce past test results by saving to a file.
By selecting and specifying various recipes during automatic defect inspection, the wafer is loaded according to the transfer recipe, the wafer is aligned on the stage according to the alignment recipe, focus conditions are set according to the focus map recipe, and inspection is performed according to the inspection recipe And the wafer is unloaded according to the transfer recipe (FIGS. 20A and 20B).

Control device 2
The defect inspection control device 2 (FIG. 6) includes a plurality of controllers as shown in FIG.
The main controller controls the GUI unit / sequence operation of the device (EBI), receives operation commands from the factory host computer or GUI, and gives necessary instructions to the VME controller and IPE controller. The main controller is provided with a man-machine interface through which operator operations are performed (various instructions / commands, recipe input, inspection start instructions, automatic and manual inspection mode switching, manual operation, etc. Input all necessary commands in inspection mode etc.). In addition, communication with the host computer in the factory, control of the evacuation system, wafer transfer, alignment control, transmission of commands to the control controller and stage controller, reception of information, and the like are also performed by the main controller. Also, acquisition of image signal from optical microscope, stage vibration correction function that feeds back stage fluctuation signal to electron optical system to correct image deterioration, wafer observation position Z-axis direction (axial direction of secondary optical system) And an automatic focus correction function for automatically correcting the focus by detecting the displacement of the lens and feeding back to the electron optical system. Transmission / reception of feedback signals and the like to the electron optical system and transmission / reception of signals from the stage apparatus are performed via the IPE controller and the stage controller, respectively.

The VME controller manages the operation of the equipment (EBI) components and gives instructions to the stage controller and PLC controller in accordance with instructions from the main controller.
The IPE controller acquires defect inspection information from the IPE node computer, classifies the acquired defects, and displays an image according to an instruction from the main controller. The IPE node computer acquires an image output from the TDI camera and performs defect inspection. Control of the electron optical system 70, that is, control of an electron gun, a lens, an aligner, etc. is also performed. For example, controlling the power supply so that a constant electron current is always irradiated even when the magnification changes to the irradiation area, automatically setting the voltage to each lens system and aligner corresponding to each magnification, etc. Control (interlocking control) such as automatic voltage setting for each lens system and aligner corresponding to each operation mode.

The PLC controller receives an instruction from the VME controller, and performs abnormality monitoring such as driving of a device such as a valve, acquisition of sensor information, and abnormality in vacuum degree that requires constant monitoring.
Upon receiving an instruction from the VME controller, the stage controller moves in the XY directions and rotates the wafer installed on the stage. In particular, the stage controller enables precise movement in the order of μm in the X-axis direction and Y-axis direction (allowable error of about ± 0.5 μm), and controls the rotation direction within an error accuracy of about ± 0.3 seconds. (Θ control) is also possible.
By configuring such a distributed control system, it is not necessary to change the software and hardware of the host controller by keeping the interface between the controllers the same when the terminal device configuration device is changed. Further, even when the sequence operation is added / modified, it is possible to flexibly cope with the configuration change by minimizing the change of the upper software and the hardware.

User Interface FIG. 23 shows a device configuration of the user interface unit.
The input unit is a device that receives input from the user, and includes a “keyboard”, “mouse”, and “joy pad”.
The display unit is a device that displays information to the user, and includes two monitors. The monitor 1 displays an image acquired by a CCD camera or a TDI camera, and the monitor 2 performs GUI display.
The progress status of the inspection may be displayed on the screen by color in real time. If the wafer position information such as which wafer is located and the information on the inspection target wafer, such as how far the inspection has been completed and where there is a defect, are displayed by color, the progress of the inspection becomes obvious at a glance. Further, the die to be inspected may be displayed for each swath.

In this apparatus, the following three coordinate systems are defined.
(1) Stage coordinate system [X _S , Y _S ]
This is a reference coordinate system for position indication during stage position control, and there is only one in this apparatus.
Using the lower left corner of the chamber as the origin, the X coordinate value increases in the right direction, and the Y coordinate value increases in the upward direction.
The position (coordinate value) indicated in the stage coordinate system is the center of the stage (wafer center). That is, when the coordinate value [0, 0] is specified in the stage coordinate system, the stage center (wafer center) moves so as to overlap the origin of the stage coordinate system.
The unit is [μm], but the minimum resolution is λ / 1024 (≈0.618 [μm]). Where λ is the wavelength of the laser used in the laser interferometer (λ≈632.991 [μm]).
(2) Wafer coordinate system [X _W , Y _W ]
This is a reference coordinate system for instructing the observation (imaging / display) position on the wafer, and there is only one in this apparatus.
With the wafer center as the origin, the X coordinate value increases in the right direction and the Y coordinate value increases in the upward direction. The position (coordinate value) indicated in the wafer coordinate system is the imaging center of the imaging device (CCD camera or TDI camera) selected at that time.
The unit is [μm], but the minimum resolution is λ / 1024 (≈0.618 [μm]). (Λ is the same as above)
(3) Die coordinate system [X _D , Y _D ]
A reference coordinate system for defining an observation (imaging / display) position in each die, and exists for each die.
With the lower left corner of each die as the origin, the X coordinate value increases in the right direction and the Y coordinate value increases in the upward direction.
The unit is [μm], but the minimum resolution is λ / 1024 (≈0.618 [μm]). (Λ is the same as above)

The dies on the wafer are numbered (numbered), and the die that serves as a reference for numbering is called the origin die. By default, the die closest to the origin of the wafer coordinate system is set as the origin die, but the position of the origin die can be selected by the user's specification.
The relationship between the coordinate value in each coordinate system and the observed (displayed) position is as shown in FIG.
Further, the relationship between the coordinates instructed by the user interface and the stage moving direction is as follows.

(1) Joystick & GUI arrow button Example in which the direction indicated by the joystick and GUI arrow button is regarded as the direction that the operator wants to see and the stage is moved in the direction opposite to the indicated direction)
Direction: Right ... Stage movement direction: Left (image moves to the left = field of view moves to the right)
Direction: Up ... Stage movement direction: Down (Image moves down = Field of view moves up)
(2) Directly inputting coordinates on the GUI The coordinates directly input on the GUI are regarded as a place the operator wants to see on the wafer coordinate system, and the stage is moved so that the corresponding wafer coordinates are displayed at the center of the captured image. .
In the apparatus described with reference to FIG. 14, a procedure is adopted in which a correction ring is placed on the electrostatic chuck and the wafer is positioned so as to be applied to the inner diameter of the correction ring. Therefore, in the inspection apparatus shown in FIG. 15, a correction ring is mounted on the wafer in the load lock chamber 22. 1, the wafer with the correction ring mounted is transported together with the correction ring and introduced into the sample chamber 21. The procedure of attaching to the upper electrostatic chuck is taken. As a mechanism for realizing this, there is an elevator mechanism shown in FIG. 25 for moving the elevator up and down to pass the wafer from the atmospheric transfer robot to the vacuum transfer robot.

Hereinafter, a procedure for carrying a wafer using this mechanism will be described.
As shown in FIG. 25A, the elevator mechanism provided in the load lock chamber has a plurality of stages (two stages in the drawing) of correction ring supports that are provided to be movable in the vertical direction. The upper correction ring support bases 22 and 2 and the lower correction ring support bases 22 and 3 are fixed to the first bases 22 and 5 that are moved up and down by the rotation of the first motors 22 and 4, thereby The rotation of the motors 22 and 4 causes the first bases 22 and 5 and the upper and lower correction ring support bases 22 and 22 and 3 to move upward or downward.

  On each correction ring support, correction rings 22 and 6 having an inner diameter corresponding to the size of the wafer are placed. Two types of correction rings 22.6 are prepared for the 200 mm wafer and for the 300 mm wafer with different inner diameters, and the outer diameters of these correction rings are the same. Thus, by using a correction ring having the same outer diameter, mutual compatibility is created, and 200 mm and 300 mm can be placed in any combination in the load lock chamber. In other words, for the line in which the 200 mm wafer and the 300 mm wafer are mixed and flowed, the upper stage is for 300 mm and the lower stage is for 200 mm, so that it is possible to flexibly cope with the inspection regardless of which wafer flows. In addition, if the same size wafer flows, the upper and lower tiers are for 200 mm or 300 mm, and the upper and lower tier wafers can be inspected alternately, so that the throughput can be improved.

The second motors 22 and 7 are mounted on the first bases 22 and 5, and the second bases 22 and 8 are attached to the second motors 22 and 7 so as to be movable up and down. Upper wafer support tables 22 and 9 and lower wafer support tables 22 and 10 are fixed to the second tables 22 and 8. As a result, when the second motors 22 and 7 are rotated, the second tables 22 and 8 and the upper and lower wafer support tables 22 and 9 and 22 and 10 are integrally moved upward or downward.
Therefore, as shown in FIG. 25A, the wafer W is loaded on the hand of the atmospheric transfer robot 21 and 2 and loaded into the load lock chamber 22 and then, as shown in FIG. The motors 22 and 7 are rotated in the first direction to move the wafer support tables 22 and 9 and 22 and 10 upward to place the wafer W on the upper wafer support tables 22 and 9. As a result, the wafer W is transferred from the atmospheric transfer robots 21 and 2 to the wafer support tables 22 and 9. Thereafter, as shown in (C), the atmospheric transfer robots 21 and 2 are moved backward, and when the backward movement of the atmospheric transfer robots 21 and 2 is completed, as shown in (D), the second motors 22 and 7 are moved to the first. The wafer support bases 22, 9, 22, 10 are moved downward by rotating in a direction opposite to the direction of. As a result, the wafer W is placed on the upper correction ring 22.

Next, as shown in (E), the hand of the vacuum transfer robot 21. 6 is put into the load lock chamber 2. In this state, the first motor 22.4 is rotated, and as shown in FIG. 2F, the first base 22.5, the upper and lower correction ring support bases 22.2, 22-3, the second motor 22 7 and the upper and lower wafer support tables 22, 9, 22, 10 are moved downward, whereby the correction rings 21, 6 and the wafer W placed on the upper wafer support tables 22, 9 are transferred to the vacuum transfer robot 21.・ It can be placed on the hand 6 and carried into the sample chambers 21 and 7.
The operation of returning the wafer, which has been inspected in the sample chambers 21 and 7, to the load lock chambers 21 and 4 is performed in the reverse procedure to the above, and is loaded onto the wafer support table by the vacuum transfer robot together with the correction ring. The wafer is transferred to the correction ring support table, then to the wafer support table, and finally placed on the atmospheric transfer robot. In FIG. 25, the wafer transfer operation in the upper stage has been described. However, the same operation is possible in the lower stage by adjusting the heights of the hands of the atmospheric transfer robots 21 and 2 and the vacuum transfer robots 21 and 6. is there. As described above, by appropriately switching the heights of the hands of the atmospheric transfer robots 21 and 2 and the vacuum transfer robots 21 and 6, an uninspected wafer is carried into the sample chamber from one stage, and the inspected wafer is then sampled. Unloading from the chamber to the other stage can be performed alternately.

Loader 60
The loader 60 (FIG. 6) includes a robot-type first transfer unit 61 disposed in the housing 22 of the mini-environment device 20 and a robot-type second transfer disposed in the second loading chamber 42. And a unit 63.
The first transport unit 61 has a multi-node arm 612 that is rotatable about the axis O ₁ -O ₁ with respect to the drive unit 611. Although an arbitrary structure can be used as a multi-node arm, in this embodiment, it has three parts rotatably attached to each other. One portion of the arm 612 of the first transport unit 61, that is, the first portion closest to the drive unit 611 is a shaft that can be rotated by a general-purpose drive mechanism (not shown) provided in the drive unit 611. 613 is attached. The arm 612 can be rotated around the axis O ₁ -O ₁ by the shaft 613 and can expand and contract in the radial direction with respect to the axis O ₁ -O ₁ as a whole by relative rotation between the portions. A gripping device 616 for gripping a wafer, such as a general-purpose mechanical chuck or electrostatic chuck, is provided at the tip of the third portion farthest from the shaft 613 of the arm 612. The drive unit 611 can be moved in the vertical direction by a lifting mechanism 615 having a general structure.

  In the first transport unit 61, the arm 612 extends in one direction M1 or M2 (FIG. 7) of the two cassettes c held in the cassette holder 10, and the inside of the cassette c The wafer W accommodated in the wafer is placed on the arm or is held and taken out by a chuck (not shown) attached to the tip of the arm. Thereafter, the arm contracts (the state shown in FIG. 7), and the arm rotates to a position where it can extend in the direction M3 of the pre-aligner 25, and stops at that position. Then, the arm extends again and the wafer W held by the arm is placed on the pre-aligner 25. After receiving the wafer from the pre-aligner 25 in the opposite direction, the arm further rotates and stops at a position where the arm can extend toward the first loading chamber 41 (direction M4), and the wafer in the first loading chamber 41 is stopped. The wafer is delivered to the receiver 47. When the wafer is mechanically gripped, the peripheral edge of the wafer (in the range of about 5 mm from the peripheral edge) is gripped. This is because a device (circuit wiring) is formed on the entire surface of the wafer except for the peripheral portion, and if a portion other than the peripheral portion is gripped, the device is broken or a defect is generated.

The second transfer unit 63 is basically the same in structure as the first transfer unit 61, and only the point that the wafer W is transferred between the wafer rack 47 and the mounting surface of the stage device 50. Is different.
The first and second transfer units 61 and 63 transfer the wafer from the cassette c held in the cassette holder onto the stage device 50 arranged in the working chamber 31 and vice versa, and the wafer is in a substantially horizontal state. Keep it in place. The arms of the transfer units 61 and 63 move up and down simply by taking out the wafer from the cassette c and inserting it into the cassette c, placing the wafer on the wafer rack and taking it out from the wafer rack, and the stage device 50. It is only necessary to place the wafer on and take it out of the wafer. Therefore, for example, even a large wafer having a diameter of 30 cm can be moved smoothly.

Here, in the inspection system 1 having the above-described configuration, the wafer transfer from the cassette c supported by the cassette holder 10 to the stage device 50 disposed in the working chamber 31 will be described in order.
As described above, the cassette holder 10 has a structure suitable for manually setting a cassette, and a cassette holder 10 having a structure suitable for automatically setting a cassette. In this embodiment, when the cassette c is set on the lifting table 11 of the cassette holder 10, the lifting table 11 is lowered by the lifting mechanism 12 and the cassette c is aligned with the entrance / exit 225. When the cassette is aligned with the entrance / exit 225, a cover (not shown) provided on the cassette c is opened, and a cylindrical cover is disposed between the cassette c and the entrance / exit 225 of the mini-environment device 20. Thus, the cassette and the mini-environment space 21 are blocked from the outside. When a shutter device that opens and closes the entrance / exit 225 is provided on the mini-environment device 20 side, the shutter device operates to open the entrance / exit 225.

On the other hand, the arm 612 of the first transport unit 61 is stopped in a state facing in either the direction M1 or M2 (in this description, the direction of M1). One of the wafers stored in the cassette c is received at the tip.
When the reception of the wafer by the arm 612 is completed, the arm contracts, the shutter device operates to close the entrance / exit (if there is a shutter device), and then the arm 612 rotates about the axis O ₁ -O _1. In this state, it can be extended in the direction M3. Then, the arm is extended and placed on the tip or held by the chuck, and the wafer is placed on the pre-aligner 25, and the orientation of the wafer in the rotation direction (direction around the central axis perpendicular to the wafer plane) by the pre-aligner. Is positioned within a predetermined range. When the positioning is completed, the first transfer unit 61 receives the wafer from the pre-aligner 25 at the tip of the arm 612 and then contracts the arm so that the arm can be extended in the direction M4. Then, the door 272 of the shutter device 27 moves to open the entrances 226 and 436 and the arm 612 extends to place the wafer on the upper stage side or the lower stage side of the wafer rack 47 in the first loading chamber 41. Note that the opening 435 formed in the partition wall 434 is closed in an airtight state by the door 461 of the shutter device 46 before the shutter device 27 is opened and the wafer is transferred to the wafer rack 47.

  During the wafer transfer process by the first transfer unit 61 described above, clean air flows in a laminar flow (as a down flow) from the gas supply unit 231 provided in the housing body 22 of the mini-environment device 20, and is being transferred. This prevents dust from adhering to the upper surface of the wafer. A part of the air around the transport unit (in this embodiment, air mainly contaminated with about 20% of the air supplied from the supply unit) is sucked from the suction duct 241 of the discharge device 24 and discharged out of the housing. Is done. The remaining air is recovered through a recovery duct 232 provided at the bottom of the housing body 22 and returned to the gas supply unit 231 again.

  When a wafer is placed on the wafer rack 47 in the first loading chamber 41 of the loader housing 40 by the first transfer unit 61, the shutter device 27 is closed and the loading chamber 41 is sealed. Then, after the air is expelled into the loading chamber 41 and filled with the inert gas, the inert gas is also discharged, and the inside of the loading chamber 41 becomes a vacuum atmosphere. The vacuum atmosphere in the loading chamber 41 may be a low degree of vacuum. When the degree of vacuum of the loading chamber 41 is obtained to some extent, the shutter device 46 operates to open the doorway 434 that has been sealed with the door 461, and then the arm 632 of the second transport unit 63 extends to hold the tip gripping device. Then, one wafer is received from the wafer receiver 47 (mounted on the tip or held by a chuck attached to the tip). When the receipt of the wafer is completed, the arm contracts, and the shutter device 46 operates again to close the doorway 435 by the door 461. Note that before the shutter device 46 is opened, the arm 632 can be extended in advance in the direction N1 of the wafer rack 47. In addition, as described above, before the shutter device 46 is opened, the doors 437 and 325 are closed by the door 452 of the shutter device 45 to prevent communication between the second loading chamber 42 and the working chamber 31. The inside of the second loading chamber 42 is evacuated.

When the shutter device 46 closes the entrance / exit 435, the second loading chamber 42 is evacuated again and is evacuated at a higher degree of vacuum than the first loading chamber 41. Meanwhile, the arm of the second transfer unit 61 is rotated to a position where it can extend toward the stage device 50 in the working chamber 31. On the other hand, in the stage apparatus 50 in the working chamber 31, the Y table 52 has an X axis line X ₁ − that passes through the rotation axis O ₂ −O ₂ of the second transport unit 63 with the center line X ₀ -X ₀ of the X table 53. It moves upward in FIG. 13 to a position that substantially coincides with X _1, and the X table 53 has moved to a position that is closest to the leftmost position in FIG. 2, and stands by in this state. When the second loading chamber 42 becomes substantially the same as the vacuum state of the working chamber 31, the door 452 of the shutter device 45 moves to open the entrances 437 and 325, the arms extend, and the tip of the arm holding the wafer is the working chamber. The stage apparatus 50 in 31 is approached. Then, the wafer W is placed on the placement surface 551 of the stage apparatus 50. When the placement of the wafer is completed, the arm contracts and the shutter device 45 closes the entrances 437 and 325.

  The operation until the wafer W in the cassette c is transferred and placed on the placement surface 551 of the stage device 50 has been described above. In order to return the wafer W that has undergone the inspection process from the stage apparatus 50 to the cassette c, an operation reverse to that described above is performed. In addition, since the plurality of wafers are placed on the wafer rack 47, the first transfer unit 63 is being transferred while the second transfer unit 63 is transferring the wafer between the wafer rack 47 and the stage device 50. However, the wafer can be transferred between the cassette c and the wafer rack 47. Therefore, the inspection process can be performed efficiently.

Precharge unit 81
As shown in FIG. 6, the precharge unit 81 is disposed adjacent to the lens barrel 71 of the electron optical system 70 in the working chamber 31. Since the inspection system 1 of the present invention is an apparatus of a type that inspects a device pattern or the like formed on the wafer surface by scanning and irradiating the wafer with an electron beam, depending on conditions such as wafer material and energy of irradiated electrons. The wafer surface may be charged (charged up). In addition, there may be places where the wafer surface is strongly charged and weakly charged. Information on secondary electrons generated by electron beam irradiation is used as information on the wafer surface. If there is unevenness in the amount of charge on the wafer surface, information on the secondary electrons will also be included and an accurate image can be obtained. I can't. Therefore, in this embodiment, a precharge unit 81 is provided to prevent uneven charging. The precharge unit 81 includes a charged particle irradiation unit 811, and eliminates uneven charging by irradiating the charged particle irradiation unit 811 with charged particles before irradiating the wafer with primary electrons for inspection. The charged state of the wafer surface can be detected by forming an image of the wafer surface in advance using the electron optical system 70 and evaluating the image, and charging based on the detected charged state. The irradiation of charged particles from the particle irradiation unit 811 is controlled. In the precharge unit 81, the primary electron beam may be blurred and irradiated.

Alignment control device 87
The alignment control device 87 is a device that positions the wafer W with respect to the electron optical system 70 using the stage device 50. The alignment control device 87 is a low-magnification alignment (alignment with a lower magnification than that of the electron optical system), which is an approximate alignment of the wafer by wide-field observation using an optical microscope 871 (FIGS. 6 and 26), an electron optical system Control of wafer high-magnification alignment, focus adjustment, inspection area setting, pattern alignment, and the like using 70 electron optical systems is performed. The wafer is inspected at such a low magnification in order to automatically inspect the wafer pattern when performing wafer alignment by observing the wafer pattern with a narrow field of view using an electron beam. This is because it is necessary to easily detect the alignment mark by the electron beam.
The optical microscope 871 is provided in the main housing 30, but may be provided to be movable in the main housing 30. A light source (not shown) for operating the optical microscope 871 is also provided in the main housing 30. Further, the electron optical system that performs high-magnification observation shares the electron optical system (the primary optical system 72 and the secondary optical system 74) of the electron optical system 70.

  FIG. 26 shows a schematic configuration of the alignment control device 87. In order to observe the observation point on the wafer W at a low magnification, the observation point on the wafer is moved into the field of view of the optical microscope by moving the X stage or the Y stage of the stage apparatus 50. The wafer is viewed with a wide field of view using the optical microscope 871, the position to be observed on the wafer is displayed on the monitor 873 via the CCD 872, and the observation position, that is, the position of the observation point is roughly determined. In this case, the magnification of the optical microscope 871 may be gradually changed from a low magnification to a high magnification.

Next, the stage device 50 is moved by a distance corresponding to the distance δx between the optical axis of the electron optical system 70 and the optical axis of the optical microscope 871, so that an observation point on the wafer determined in advance using the optical microscope 871 is used. Is moved to the visual field position of the electron optical system 70. In this case, the distance between the axis O ₃ -O ₃ of the electron optical system 70 and the optical axis O ₄ -O ₄ of the optical microscope 871 (in this embodiment, both are displaced only in the X-axis direction). (However, it may be displaced in the Y-axis direction.) Since δx is known in advance, the observation point can be moved to the visual recognition position by moving the value δx. After the movement of the observation point to the visual recognition position of the electron optical system 70 is completed, the observation point is imaged by SEM at a high magnification by the electron optical system, and an image is stored or displayed on the monitor 765.

In this way, after the observation point of the wafer is displayed on the monitor at a high magnification by the electron optical system 70, the positional deviation in the rotation direction of the wafer with respect to the rotation center of the rotary table 54 of the stage device 50, that is, a known method. A deviation δθ in the rotation direction of the wafer with respect to the optical axis O ₃ -O ₃ of the electron optical system is detected, and a positional deviation of a predetermined pattern related to the electron optical system 70 in the X axis and Y axis directions is detected. Then, the wafer alignment is performed by controlling the operation of the stage device 50 based on the detected value and the data of the inspection mark provided on the wafer obtained separately or the data on the pattern shape of the wafer.

The alignment procedure will be described in more detail.
The arrangement direction of the die of the wafer loaded on the stage does not necessarily coincide with the scanning direction of the TDI camera (see FIG. 27). In order to make this coincide, an operation of rotating the wafer on the θ stage is required, and this operation is called alignment (FIG. 28). In the alignment recipe, the alignment execution condition after being loaded on the stage is stored.
A die map (FIG. 29) showing the arrangement of dies is created at the time of alignment, and the die size, the position of the origin die (which is the starting point showing the position of the die), etc. are stored in the die map recipe.

As an alignment (positioning) procedure, first, coarse positioning is performed at a low magnification of the optical microscope, then detailed positioning is performed at the high magnification of the optical microscope, and finally by an EB image.
A. Imaging at low magnification using an optical microscope (1) <Specifying first, second, third search die and specifying template>
(1-1) First Search Die Designation and Template Designation The user moves the stage so that the lower left corner of the die located below the wafer is located near the center of the camera. get. This die is a die serving as a positioning reference, and the coordinates of the lower left corner are the coordinates of the feature points. In the future, the exact position coordinates of any die on the substrate will be measured by pattern matching with this template image. As the template image, an image that has a unique pattern within the search area must be selected.
In this embodiment, the lower left corner is the pattern matching template image acquisition position. However, the present invention is not limited to this, and an arbitrary position in the die may be selected as the feature point. However, in general, it is preferable to select one of the four corners because the corners are easier to specify the coordinates than the points inside the die or on the sides. Similarly, in the present embodiment, a pattern matching template image is acquired for a die located below the wafer, but it is natural that any die may be selected so that alignment can be easily performed.

(1-2) Second search die designation The die next to the right of the first search die is the second search die, and the stage is moved by the user operation so that the lower left corner of the second search die is located near the center of the camera. After the position is determined, pattern matching is automatically performed using the template image acquired in (1-1) above, so that the pattern of the second search die that matches the template image specified by the first search die Get the coordinate value.
In the present embodiment, the die adjacent to the right of the first search die has been described as an example of the second search die. However, the second search die of the present invention is not limited to this. . In short, it is only necessary to select a point at which the positional relationship of the die in the row direction can be accurately grasped by pattern matching from the reference point where the position map of the accurate feature point is grasped. Therefore, for example, the die next to the left of the first search die can be used as the second search die.

(1-3) Third search die designation The die next to the second search die is the third search die, and the stage is moved by the user operation so that the lower left corner of the third search die is located near the center of the camera. After the position is determined, pattern matching is automatically performed using the template image acquired in (1-1) above, so that the pattern of the third search die that matches the template image specified by the first search die Get the coordinate value.
In the present embodiment, the upper die adjacent to the second search die is described as an example of the third search die. However, it goes without saying that the third search die of the present invention is not limited to this. In short, it is only necessary to be able to grasp the positional relationship including the distance of the coordinates of the specific point of the die in the column direction with reference to the die that grasped the exact coordinates of the feature points. Therefore, the die adjacent to the upper side of the first search die can be preferably applied as an alternative.

(2) <Light microscope low magnification Y direction pattern matching>
(2-1) From the relationship between the pattern match coordinates (X2, Y2) of the second search die and the pattern match coordinates (X3, Y3) of the third search die (dX, dY) Is calculated.
dX = X3-X2
dY = Y3-Y2
(2-2) Using the calculated movement amount (dX, dY), move the stage to the coordinates (XN, YN) where the pattern of the adjacent die on the first search die exists (expected).
XN = X1 + dX
YN = Y1 + dY
However, (X1, Y1) is currently being observed by taking the coordinates of the pattern of the first search die (2-3), moving the stage, taking an image at low magnification, and performing pattern matching using the template image. The exact coordinate values (XN, YN) of the pattern are acquired, and 1 is set as the initial value of the number of detected die (DN).
(2-4) The amount of movement (dX, dY) from the pattern coordinates (X1, Y1) of the first search die to the coordinates (XN, YN) of the pattern currently being imaged is calculated.
dX = XN-X1
dY = YN-Y1
(2-5) The stage is moved from the first search die as a starting point by a moving amount (2 * dX, 2 * dY) twice the calculated moving amount (dX, dY).
(2-6) After moving the stage, the image is captured at low magnification, and pattern matching is executed using the template image, thereby updating the exact coordinate value (XN, YN) of the currently observed pattern, The number of detected signals is doubled. See FIG. 30 for this.
(2-7) Steps (2-4) to (2-6) are repeatedly executed toward the upper portion of the wafer until the Y coordinate value specified in advance is exceeded.

  In the present embodiment, an example has been described in which a double movement amount is repeated in order to increase accuracy, reduce the number of times of processing (number of repetitions), and shorten the processing time. However, there is a problem in accuracy. If the processing time is to be further reduced, it may be executed at a high magnification of an integral multiple such as 2 or more, such as 3 or 4 times. Conversely, if there is no problem, the movement may be repeated with a fixed movement amount in order to further improve the accuracy. In any of these cases, it goes without saying that this is also reflected in the detected number.

(3) <Light microscope low magnification θ rotation>
(3-1) The amount of movement from the pattern coordinate (X1, Y1) of the first search die to the exact coordinate value (XN, YN) of the last searched die pattern and the number of dies detected so far ( DN) is used to calculate the rotation amount (θ) and the Y-direction die size (YD) (see FIG. 31).
dX = XN-X1
dY = YN-Y1
θ = tan ⁻¹ (dX / dY)
YD = √ ((dX) ² + (dY) ² ) / DN
(3-2) The θ stage is rotated by the calculated rotation amount (θ).
B. Imaging at high magnification using an optical microscope (1) The same procedure as (1) for light microscope low magnification is performed using a light microscope high magnification image.
(2) The same procedure as in (2) for light microscope low magnification is executed using a light microscope high magnification image.
(3) The same procedure as (3) for light microscopic low magnification is executed.

(4) <Tolerance value check after optical microscope magnification θ rotation>
(4-1) [First search die, light magnifying power template specification]
The coordinates (X′1, Y′1) of the first search die after rotation are calculated from the coordinates (X1, Y1) before rotation and the rotation amount (θ), and the stage is moved to the coordinates (X′1, Y′1). After moving and locating, get a template image for pattern matching.
X′1 = x ₁ * cos θ−y ₁ * sin θ
Y′1 = x ₁ * sin θ + y ₁ * cos θ
(4-2) Light-magnification high-magnification Y-direction pattern matching The pattern currently being observed by moving in the Y direction by dY from the coordinates (X′1, Y′1) of the first search die after rotation and executing pattern matching The exact coordinate values (XN, YN) are obtained.
(4-3) The amount of movement (dX, dY) from the coordinates (X′1, Y′1) of the first search die after rotation to the coordinates (XN, YN) of the pattern currently being imaged is calculated.
dX = XN−X′1
dY = YN−Y′1
(4-4) The stage is moved from the first search die as a starting point by a movement amount (2 * dX, 2 * dY) twice the calculated movement amount (dX, dY).
(4-5) After moving the stage, imaging is performed at an optical microscope magnification, and pattern matching is executed using a template image, thereby updating the exact coordinate values (XN, YN) of the currently observed pattern.
(4-6) Steps (4-3) to (4-5) are repeatedly executed toward the upper portion of the wafer until the Y coordinate value designated in advance is exceeded.
(4-7) Calculate the rotation amount of θ Move from the coordinate (X′1, Y′1) of the first search die after rotation to the exact coordinate value (XN, YN) of the die pattern searched last The rotation amount (θ) is calculated using the amount.
dX = XN-X1
dY = YN-Y1
θ = tan ⁻¹ (dX / dY)
(4-8) It is confirmed that the rotation amount (θ) calculated in the optical sensible height θ allowable value check (4-7) is within a predetermined value or less. If not, after the θ stage is rotated using the calculated rotation amount (θ), (4-1) to (4-8) are executed again. However, if (4-1) to (4-8) are repeatedly executed a prescribed number of times and the result does not fall within the allowable range, the processing is interrupted as an error.

C. EB image alignment (1) <Y search first die, EB template designation>
The same procedure as that in (1) for optical magnification is performed using the EB image.
(2) <EB Y-direction pattern matching>
A procedure similar to that in (2) of the optical magnification is executed using the EB image.
(3) <EB θ rotation>
A procedure similar to that in (3) of the optical magnification is performed using the EB image.
(4) <Allowable value check after EB θ rotation>
A procedure similar to that in (4) of the optical magnification is performed using the EB image.
(5) If necessary, execute (1) to (4) using a high-magnification EB image. (6) First search die coordinates (X1, Y1) and second search die coordinates (X2, Y2). ) To calculate the approximate value of the X direction die size (XD) dX = X2-X1
dY = Y2-Y1
XD = √ ((dX) ² + (dY) ² )

D. Die map recipe creation (1) <X search first die, EB template designation>
The stage is moved by the user operation so that the lower left corner of the die located at the left end of the wafer is located near the center of the TDI camera, and after determining the position, a pattern matching template image is acquired. As the template image, an image that has a unique pattern within the search area must be selected.

(2) <EB X direction pattern matching>
(2-1) Using the approximate X-direction die size (XD), move the stage to the coordinates (X1 + XD, Y1) where the pattern of the die next to the X search first die exists (and is expected).
(2-2) After moving the stage, an EB image is taken with a TDI camera, and pattern matching is executed using a template image to obtain exact coordinate values (XN, YN) of the currently observed pattern, Further, 1 is set as the initial value of the number of detected die (DN).
(2-3) X Search The movement amount (dX, dY) from the pattern coordinates (X1, Y1) of the first die to the coordinates (XN, YN) of the pattern currently being imaged is calculated.
dX = XN-X1
dY = YN-Y1
(2-4) The stage is moved starting from the X search first die by a movement amount (2 * dX, 2 * dY) twice the calculated movement amount (dX, dY). (2-5) Stage movement After that, an EB image is captured by a TDI camera, and pattern matching is executed using a template image, thereby updating the exact coordinate values (XN, YN) of the currently observed pattern and reducing the number of detected dies to 2 Double.
(2-6) The steps (2-3) to (2-5) are repeatedly executed in the right direction of the wafer until the X coordinate value designated in advance is exceeded.

(3) <Calculate X-direction tilt>
The amount of movement from the X search first die pattern coordinates (X1, Y1) to the exact coordinate value (XN, YN) of the last searched die pattern and the number of dies detected so far (DN) are used. The stage direct error (Φ) and the X-direction die size (XD) are calculated.
dX = XN-X1
dY = YN-Y1
Φ = tan ⁻¹ (dY / dX)
XD = √ ((dX) ² + (dY) ² ) / DN

(4) <Die map creation>
In this way, the X direction die size (XD) is obtained, and the die map (ideal die arrangement information) is created together with the Y direction die size (YD) obtained when the rotation amount (θ) is calculated in advance. To do. The die map shows the ideal die placement. On the other hand, dies on different substrates are affected by, for example, mechanical errors of the stage (components such as guides and assembly errors), errors of the interferometer (for example, problems of assembly of mirrors, etc.) and image distortion due to charge-up. However, it is not always possible to observe the available layout, but it is necessary to grasp the error between the actual die position and the ideal layout on the die map, and take this error into account. Make inspections while making corrections.

E. Focus Recipe Creation Procedure Next, a focus recipe creation procedure will be described. The focus recipe stores information on an optimum focus position at a mark position on a plane of a sample such as a substrate or various conditions related to the focus position in a predetermined format such as a table. In the focus map recipe, focus conditions are set only at specified positions on the wafer, and the focus value between the specified positions is linearly complemented (see FIG. 32). The focus recipe creation procedure is as follows.
(1) Select a focus measurement target die from the die map. (2) Set a focus measurement point in the die. (3) Move the stage to each measurement point, and focus value (CL12) based on the image and contrast value. Adjust the voltage manually.
The die map created by the alignment process is ideal position information calculated from the die coordinates at both ends of the wafer, and an error occurs between the die position on the die map and the actual die position due to various factors. (Refer to FIG. 33) A procedure for creating a parameter for absorbing this error is called fine alignment. The fine alignment recipe includes error information between a die map (ideal die arrangement information) and an actual die position. Is saved. The information set here is used at the time of defect inspection. In the fine alignment recipe, the error is measured only for the die designated on the die map, and the error between the designated dies is linearly supplemented.

F. Fine alignment procedure (1) Specify error measurement target die for fine alignment from die map (2) Select reference die from error measurement target die, and set the position of this die as the point where error with die map is zero ( 3) Capture the lower left corner of the reference die with a TDI camera to obtain a pattern matching template image (however, a unique pattern is selected as a template image in the search area)
(4) The coordinates (X0, Y0) at the lower left (on the die map) of the neighboring error measurement target die are acquired, and the stage is moved. After the movement, an image is taken with a TDI camera, and pattern matching is executed using the template image of (3), thereby obtaining exact coordinate values (X, Y).
(5) Save the error between the coordinate value (X, Y) acquired by pattern matching and the coordinate value (X0, Y0) on the die map. (6) Execute (4) to (5) for all error measurement dies. To do.

Defect detection for processing the data obtained by the electron beam apparatus according to the present invention, that is, the electron optical system 70 to acquire image data and detecting defects on the semiconductor wafer based on the obtained image data will be further described.
In general, an inspection apparatus using an electron beam, that is, an electron optical system 70 is expensive and has a lower throughput than other process apparatuses. Therefore, after an important process (for example, etching, film formation, or CMP (Chemical Mechanical Polishing) flattening process, etc.) which is considered to be most inspected at present, in the wiring process, a finer wiring process part, In other words, it is used in the first and second wiring processes, the previous gate wiring process, and the like. In particular, it is important to find a shape defect or an electrical defect such as a wiring having a line width of 100 nm or less, that is, a design rule of 100 nm or less, or a via hole having a diameter of 100 nm or less, and to feed back to the process.

As described above, the wafer to be inspected is positioned on the ultra-precision stage device (XY stage) 50 through the atmospheric transfer system and the vacuum transfer system, and then fixed by the electrostatic chuck mechanism or the like. In the defect inspection process, the position of each die and the height of each location are detected and stored by an optical microscope as necessary. In addition to this, the optical microscope obtains an optical microscope image of a desired location such as a defect and is used for comparison with an electron beam image. Next, the conditions of the electron optical system are set, and the information set by the optical microscope is corrected using the electron beam image to improve the accuracy.
Next, recipe information corresponding to the type of wafer (after which process, whether the wafer size is 200 mm or 300 mm, etc.) is input to the apparatus. After performing the above, defect inspection is usually performed in real time while acquiring an image. Cell-to-cell comparison, die comparison, and the like are inspected by a high-speed information processing system equipped with an algorithm, and the results are output to a CRT or the like or stored in a memory as necessary.

  The basic flow of defect inspection is shown in FIG. First, after carrying the wafer including the alignment operation 113. 1, a recipe in which conditions relating to the inspection are set is created (113. 2). At least one type of recipe is required for the wafer to be inspected, but there may be a plurality of recipes for one wafer to be inspected in order to cope with a plurality of inspection conditions. Further, when there are a plurality of wafers to be inspected with the same pattern, a plurality of wafers may be inspected with one type of recipe. The path 1113 in FIG. 34 indicates that when an inspection is performed with a recipe created in the past, it is not necessary to create a recipe immediately before the inspection operation.

In FIG. 34, the inspection operation 1113 performs a wafer inspection according to the conditions and sequence described in the recipe. The defect extraction is performed immediately every time a defect is found during the inspection operation, and the following operations are executed in almost parallel.
・ Operation to perform defect classification (113/5) and add extracted defect information and defect classification information to result output file ・ Operation to add extracted defect image to image-only result output file or file ・ Defects such as position of extracted defect Operation for Displaying Information on Operation Screen When inspection is completed in units of wafers to be inspected, the following operations are then executed in substantially parallel.
・ Operation to close and save result output file ・ Operation to send inspection result when communication from outside requests inspection result ・ Operation to eject wafer When next wafer is set to be inspected continuously The wafer to be inspected is transferred, and the above series of operations is repeated.

In creating a recipe in FIG. 34, a recipe including a setting file for conditions related to inspection is created. The recipe can be stored, and conditions are set at the time of inspection or before inspection using the recipe. Conditions relating to the inspection described in the recipe include, for example, the following matters.
・ Die to be inspected ・ Die internal inspection area ・ Inspection algorithm ・ Detection conditions (Inspection sensitivity and other conditions necessary for defect extraction)
・ Observation conditions (magnification, lens voltage, stage speed, inspection order, etc.)

Among the inspection conditions described above, the inspection target die is set by the operator specifying a die to be inspected on the die map screen displayed on the operation screen as shown in FIG. In the example of FIG. 35, the die 1 on the wafer end face and the die 2 that is clearly determined to be defective in the previous process are grayed out and deleted from the inspection target, and the rest are used as the inspection target die. It also has a function of automatically designating an inspection die based on the distance from the wafer end face and the quality information of the die detected in the previous process.
Also, the setting of the inspection area inside the die is based on the image acquired by the operator using an optical microscope or EB microscope with respect to the inspection area setting screen displayed on the operation screen as shown in FIG. And using an input device such as a mouse. In the example of FIG. 36, a region 115 · 1 indicated by a solid line and a region 115 · 2 indicated by a broken line are set.

In the region 115. 1, almost the entire die is set as a setting region. In this case, the inspection algorithm is an adjacent die comparison method, and details of detection conditions and observation conditions for this region are set separately. In the area 115. 2, the inspection algorithm is an array inspection, and details of detection conditions and observation conditions for this area are set separately. That is, a plurality of inspection areas can be set, and each inspection area can be set with its own inspection algorithm and inspection sensitivity. In addition, the inspection areas can be overlapped, and different inspection algorithms can be simultaneously processed for the same area.
In the inspection operation 113.4 in FIG. 34, as shown in FIG. 37, the wafer to be inspected is divided into scan widths and scanned. The scanning width is substantially determined by the length of the line sensor, but is set so that the end portions of the line sensor slightly overlap. This is to ensure a margin for image alignment when determining the continuity between lines or when performing a comparative inspection when finally integrating the detected defects. The amount of overlap is about 16 dots for a 2048-dot line sensor.

FIG. 38 schematically shows the scanning direction and sequence. That is, in order to shorten the inspection time, a bidirectional operation (A operation) and a unidirectional operation (B operation) due to machine limitations can be selected by the operator.
In addition, it has a function of automatically calculating and inspecting an operation for reducing the scanning amount based on the inspection target die setting of the recipe. FIG. 39A shows an example of scanning when there is only one inspection die, and unnecessary scanning is not performed.
The inspection algorithm set by the recipe can be roughly divided into cell inspection (array inspection) and die inspection (random inspection).
As shown in FIG. 39 (B), the die is divided into a cell portion 118.2 having a periodic structure mainly used for a memory and a random portion 118.3 having no periodic structure. Since there are a plurality of comparison objects in the same die, the cell portions 118 and 2 having a periodic structure can be inspected by cell inspection, that is, by comparing cells in the same die. On the other hand, since the random parts 118 and 3 do not have a comparison target in the same die, it is necessary to compare the dies by die inspection.

The die inspection is further classified as follows according to the comparison target.
・ Adjacent die comparison method (Die-to-Die inspection)
・ Standard die comparison method (Die-to-Any Die inspection)
・ CAD data comparison method (Cad Data-to-Any Die inspection)
A method generally called a golden template method is a standard die comparison method and a CAD data comparison method. In the standard die comparison method, a reference die is a golden template, and in a CAD data comparison method, CAD data is a golden template. .

The operation of each inspection algorithm is described below.
Cell inspection (array inspection)
Cell inspection is applied to inspection of periodic structures. An example is a DRAM cell.
In the inspection, the reference image to be referred to is compared with the image to be inspected, and the difference is extracted as a defect. The reference image and the image to be inspected may be a binary image or a multi-value image in order to improve detection accuracy.
The defect may be the difference itself between the reference image and the image to be inspected, but it is a secondary determination for preventing erroneous detection based on difference information such as the difference amount of the detected difference and the total area of the pixels having the difference. May be performed.
In the cell inspection, the comparison between the reference image and the image to be inspected is performed in units of structure periods. That is, comparison may be made in units of one structure period while reading out images collectively acquired by a CCD or the like. If the reference image is n structure period units, n structure period units can be compared simultaneously.
An example of a method for generating a reference image is shown in FIG. 40. Here, an example in which comparison is made in units of one structure period will be described, and thus one structure period unit generation is shown. It is also possible to set the number of periods to n in the same way.
As a premise, the inspection direction in FIG. Period 4 is the inspection period. Since the operator inputs the magnitude of the period while viewing the image, periods 1 to 6 in FIG. 40 can be easily recognized.
The reference period image is generated by adding and averaging the periods 1 to 3 immediately before the inspection period in each pixel. Even if a defect exists in any one of 1 to 3, the average process is performed, so the influence is small. A defect is extracted by comparing the formed reference periodic image and the inspection periodic image 4.
Next, when inspecting the periodic image 5 to be inspected, the reference periodic images are generated by averaging the periods 2 to 4. In the same manner, a periodic image to be inspected is generated from images obtained before acquiring the periodic image to be inspected, and the inspection is continued.

Die inspection (random inspection)
The die inspection can be applied without being limited to the die structure. In the inspection, the reference image to be referred to is compared with the image to be inspected, and the difference is extracted as a defect. The reference image and the image to be inspected may be a binary image or a multi-value image in order to improve detection accuracy. The defect may be the difference between the reference image and the image to be inspected, but based on difference information such as the difference amount of the detected difference and the total area of the pixels having the difference, a secondary determination is made to prevent erroneous detection. You can go. Die inspection can be classified according to how to obtain a reference image. The operations of the adjacent die comparison method, the reference die comparison inspection method, and the CAD data comparison method included in the die inspection will be described below.

A. Adjacent die comparison method (Die-Die inspection)
The reference image is a die adjacent to the image to be inspected. A defect is judged by comparing with two dies adjacent to the image to be inspected. That is, in FIG. 41 and FIG. 42, the switches 121 and 4 and the switches 121 and 5 are set so that the memories 121 and 1 and the memories 121 and 2 of the image processing apparatus are connected to the paths 121 and 41 from the cameras 121 and 3. Has the following steps.
a) A step of storing the die image 1 in the memory 121 · 1 from the path 121 · 41 according to the scanning direction S.
b) A step of storing the die image 2 in the memory 121 · 2 from the route 121 · 41.
c) While obtaining the die image 2 from the paths 121 and 42 simultaneously with the above b), the obtained die image 2 is compared with the image data stored in the memory 121 · 1 having the same relative position on the die and the difference is obtained. Step to seek.
d) A step of storing the difference of c).
e) A step of storing the die image 3 from the path 121 · 41 into the memory 121 · 1.
f) While obtaining the die image 3 from the paths 121 and 42 simultaneously with the above e), the obtained die image 3 is compared with the image data stored in the memory 121 and the relative position of the die is the same, and the difference is obtained. Step to seek.
g) A step of storing the difference of f) above.
h) A step of determining a defect of the die image 2 from the results stored in the above d) and g).
i) Repeat steps a) to h) in subsequent dies.
Depending on the setting, before obtaining the difference in c) and f) above, the position alignment of the two images to be compared: correction is made so that there is no position difference. Or density alignment: Correct so that there is no density difference. Alternatively, both processes may be performed.

B. Standard die comparison method (Die-Any Die inspection)
The reference die is designated by the operator. The standard die is a die existing on the wafer or a die image stored before the inspection. First, the standard die is scanned or transferred, and the image is stored in a memory to be a reference image. That is, the steps described below with reference to FIGS. 42 and 43 are executed.
a) The operator selects a reference die from a die of a wafer to be inspected or a die image stored before inspection.
b) When the reference die is present on the wafer to be inspected, the switches 121.4, 121 are connected so that at least one of the memory 121.1 or the memory 121.2 of the image processing apparatus is connected to the path 121.41 from the camera 121.3. Setting the switches 121 and 5;
c) When the standard die is a die image stored before the inspection, at least one of the memory 121 · 1 or the memory 121 · 2 of the image processing apparatus stores the reference image that is a die image from the memory 121 · 6. A step of setting the switches 121 and 4 and the switches 121 and 5 to be connected to the paths 121 and
d) If the standard die is present on the wafer to be inspected, scanning the standard die and transferring a reference image, which is a standard die image, to the memory of the image processing apparatus.
e) A step of transferring a reference image, which is a standard die image, to the memory of the image processing apparatus without scanning, when the standard die is a die image stored before inspection.
f) A step of obtaining a difference by comparing an image obtained by sequentially scanning an image to be inspected, an image in a memory to which a reference image as a standard die image is transferred, and image data having the same relative position on the die.
g) A step of determining a defect from the difference obtained in f) above.
h) Subsequently, as shown in FIG. 50, the same part is inspected for the entire wafer with respect to the scanning position of the reference die and the die origin of the die to be inspected, and the scanning position of the reference die is changed until the entire die is inspected. While repeating d) to g) above.
Depending on the settings, before obtaining the difference in f) above, the position alignment of the two images to be compared is corrected so as to eliminate the position difference. Or density alignment: correction is made so that the density difference is eliminated. Alternatively, both processes may be performed.
The reference die image stored in the memory of the image processing apparatus in the above d) or e) may be all the reference dies or may be inspected while being updated as a part of the reference die.

C. CAD data comparison method (CAD Data-Any Die inspection)
A reference image is created from CAD data, which is an output of a semiconductor pattern design process by CAD, and used as a standard image. The reference image may be an entire die or a partial object including an inspection part.
The CAD data is usually vector data, and cannot be used as a reference image unless converted into raster data equivalent to image data obtained by a scanning operation. In this way, the following conversion is performed for the CAD data processing operation.
a) Vector data as CAD data is converted into raster data.
b) The above a) is performed in units of image scanning width obtained by scanning the inspection die during inspection.
c) In the above b), image data whose relative position on the die is the same as the image scheduled to be scanned by the die to be inspected is converted.
d) The above c) is performed by overlapping the inspection scan and the conversion work.
The above a) to d) are examples in which conversion in units of image scanning width is performed for speeding up, but inspection is possible without fixing the conversion units to the image scanning width.

In addition, at least one of the following functions is added as an additional function to the work of converting vector data into raster data.
a) Multi-value function of raster data.
b) A function of setting the multi-value gradation weight and offset in consideration of the sensitivity of the inspection apparatus with respect to a).
c) A function of performing image processing for processing pixels such as expansion and contraction after vector data is converted into raster data.

The inspection step by the CAD data comparison method performed in the apparatus of FIG. 42 is shown.
a) A step of converting CAD data into raster data by the computer 1 and generating a reference image by the additional function and storing it in the memory 121/6.
b) A step of setting the switches 121 and 4 and the switches 121 and 5 so that at least one of the memory 121 and the memory 121 and 2 of the image processing apparatus is connected to the path 121 and 7 from the memory 121 and 6.
c) A step of transferring the reference image in the memory 121.6 to the memory of the image processing apparatus.
d) A step of comparing the image obtained by sequentially scanning the image to be inspected, the image in the memory to which the reference image is transferred, and the image data having the same relative position on the die to obtain a difference.
e) A step of determining a defect from the difference obtained in d) above.
f) Subsequently, as shown in FIG. 44, the scanning position of the standard die is used as a reference image, the same portion of the die to be inspected is inspected on the entire wafer, and the above-mentioned a ) To e).
Depending on the setting, before obtaining the difference in d), correction is made so that the positional alignment of the two images to be compared, that is, the positional difference is eliminated. Alternatively, density alignment, that is, correction is made so that there is no density difference. Alternatively, both processes may be performed. Further, the reference die image stored in the memory of the image processing apparatus in c) may be all of the reference dies, or may be inspected while being updated as a part of the reference die.

So far, the algorithm of the array inspection (cell inspection) for inspecting the periodic structure and the random inspection has been described, but it is also possible to perform the cell inspection and the die inspection at the same time. In other words, the cell portion and the random portion are processed separately, and the cell portion compares the cells in the die, and at the same time, the random portion compares the adjacent die, reference die, or CAD data. In this way, the inspection time can be greatly shortened and the throughput is improved.
In this case, it is preferable that the cell portion inspection circuit is provided separately and independently. If inspection is not performed at the same time, it is possible to have one inspection circuit, set the software for cell inspection and random inspection to be switchable, and execute comparative inspection by switching software. is there. In other words, when pattern inspection is processed by applying a plurality of processing algorithms, these algorithms may be prepared separately and processed simultaneously, or an algorithm corresponding to them may be provided to provide a single circuit. You may make it process by switching by. In any case, there are a plurality of types of cell parts, which are also applicable to the case where comparison is made between cells and leather is further formed between dies or CAD data with respect to a random part.

The basic flow of the focus function is shown in FIG. First, after carrying the wafer including the alignment operation, a recipe in which conditions related to the inspection are set is created. As one of the recipes, there is a focus map recipe, and autofocus is performed during the inspection operation and the review operation according to the focus information set here. The focus map recipe creation procedure and autofocus operation procedure will be described below.
In the following example, the focus map recipe has an independent input screen, and the operator creates the focus recipe by executing the following steps. Such an input screen may be added to an input screen provided for another purpose.
a) A step of inputting focus map coordinates such as a die position for inputting a focus value and a pattern in the die. Switch 126.1 in FIG.
b) A step of setting a die pattern necessary for automatically measuring the focus value. This step can be skipped if the focus value is not automatically measured.
c) A step of setting the best focus value of the focus map coordinates determined in the above a).
Among them, in step a), the operator can designate an arbitrary die, but settings such as selection of all dies and selection of every n dies are also possible. The input screen can be selected by the operator, whether it is a diagram schematically representing the die arrangement on the wafer, or an image using an actual image.
Among these, in the step c), the operator manually sets the focus switch 126.2 in conjunction with the voltage value of the focus electrode (switch 126.3 in FIG. 46). Selection and setting are made in a mode mode (switches 126 and 4 in FIG. 46) for automatically obtaining a focus value.

The procedure for automatically obtaining the focus value in the step c) is, for example, in FIG. 47. a) Obtaining an image at the focus position Z = 1 and calculating the contrast.
b) The above a) is also performed with Z = 2, 3, and 4.
c) Regression from the contrast values obtained in a) and b) above to obtain a contrast function (FIG. 48).
d) Z which obtains the maximum value of the contrast function is obtained by calculation, and this is set as the best focus value.
For example, the die pattern necessary for automatically measuring the focus value shows good results when a line and space as shown in FIG. 48 is selected, but the contrast can be measured regardless of the shape if there is a monochrome pattern.
The best focus value of one point can be obtained by performing steps a) to d). The data format at this time is (X, Y, Z) X, Y: coordinates for obtaining focus, Z: set of best focus values, and the number of focus map coordinates (X, Y, Z determined by the focus map recipe) ) Will exist. This is called a focus map file in a part of the focus map recipe.

A method of setting the focus to the best focus during the inspection operation and the review operation for acquiring an image from the focus map recipe is performed in the following steps.
a) The position information is further subdivided based on the focus map file 1 created at the time of creating the focus map recipe, the best focus at this time is obtained by calculation, and the subdivided focus map file 2 is created.
b) The calculation of a) is performed with an interpolation function.
c) The interpolation function of b) is specified by the operator when creating a focus map recipe by linear interpolation, spline interpolation, or the like.
d) The stage XY position is monitored, and the voltage of the focus electrode is changed to the focus value described in the focus map file 2 suitable for the current XY position.

More specifically, in FIG. 49, the black circle is the focus value of the focus map file 1, and the white circle is the focus value of the focus map file 2.
1. Interpolation is performed between the focus values of the focus map file using the focus value of the focus map file.
2. The best focus is maintained by changing the focus position Z according to the scanning. At this time, the value is held until the next change position during the focus map file (white circle).

FIG. 50 shows an example of a semiconductor manufacturing plant using the electron beam apparatus according to the present invention. In FIG. 50, the electron beam apparatus is denoted by reference numeral 171.1, and information such as the lot number of the wafer to be inspected by the apparatus and the history of the manufacturing apparatus through manufacturing is provided in the SMIF or FOUP 171-2. The lot number can be read out from the memory or the lot number can be recognized by reading the SMIF, FOUP 171-2, or wafer cassette ID number. During the transfer of the wafer, the amount of moisture is controlled to prevent the metal wiring from being oxidized.
The PC 171.6 for defect detection control of the defect inspection apparatus 171.1 is connected to the information communication network 171/3 of the production line, and the production line that controls the production line via this network 171/3. Information such as the lot number of the wafer to be inspected and the inspection result can be sent to the control computer 171, 4 and each of the manufacturing apparatuses 171, 5 and another inspection system. The manufacturing apparatus 171/5 includes lithography-related apparatuses such as exposure apparatuses, coaters, curing apparatuses, developers, etc., film forming apparatuses such as etching apparatuses, sputtering apparatuses, and CVD apparatuses, CMP apparatuses, various measuring apparatuses, and other inspection apparatuses. Etc. are included.

  Although preferred embodiments of the present invention have been described in detail above, it will be apparent that modifications and changes can be made to these embodiments without departing from the technical idea of the present invention.

It is explanatory drawing which shows the electron optical system of the electron beam apparatus of the 1st Embodiment of this invention. It is sectional drawing of the Wien filter in the electron beam apparatus shown in FIG. It is explanatory drawing which shows the electron optical system of the electron beam apparatus of the 2nd Embodiment of this invention. It is explanatory drawing which shows the electron optical system of the electron beam apparatus of the 3rd Embodiment of this invention. It is explanatory drawing which shows the optical path in the electron beam apparatus shown in FIG. It is a graph for demonstrating the effect by removing an axial chromatic aberration in a mapping projection type electron beam apparatus. It is explanatory drawing which shows the defect detection system of the sample which can apply the electron beam apparatus of this invention. It is the top view of the main components of the inspection system shown in Drawing 6, and is a figure seen along line BB of Drawing 6. It is a figure which shows the relationship between the wafer conveyance box and loader of the inspection system shown in FIG. It is sectional drawing which shows the mini environment apparatus of the test | inspection system shown in FIG. 6, Comprising: It is the figure seen along line CC of FIG. FIG. 8 is a view showing a loader housing of the inspection system shown in FIG. 6, viewed along line DD in FIG. 7. It is a figure explaining the electrostatic chuck used for the inspection system concerning the present invention. It is a figure explaining the other example of the electrostatic chuck used for the test | inspection system which concerns on this invention. It is a figure explaining the further another example of the electrostatic chuck used for the test | inspection system which concerns on this invention. It is a figure explaining the bridge tool used for the inspection system concerning the present invention. It is a figure explaining other examples of the bridge tool used for the inspection system concerning the present invention. It is a figure explaining the defect inspection procedure in the electron beam apparatus which concerns on this invention. It is a figure explaining the defect inspection procedure in the electron beam apparatus which concerns on this invention. It is a figure explaining the defect inspection procedure in the electron beam apparatus which concerns on this invention. It is a figure explaining the defect inspection procedure in the electron beam apparatus which concerns on this invention. It is a figure explaining the defect inspection procedure in the electron beam apparatus which concerns on this invention. It is a figure explaining the defect inspection procedure in the electron beam apparatus which concerns on this invention. It is a figure explaining the structure of the control system in the test | inspection system which concerns on this invention. It is a figure explaining the structure of the user interface in the test | inspection system which concerns on this invention. It is a figure explaining the structure of the user interface in the electron beam apparatus which concerns on this invention. It is a figure explaining the structure and operation | movement procedure of an elevator mechanism in the load lock chamber of FIG. It is a schematic explanatory drawing of the wafer alignment control apparatus applicable to the electron optical system of the inspection system which concerns on this invention. It is a figure for demonstrating the reason for which alignment of a wafer is required. It is a figure for demonstrating the alignment execution state of a wafer. It is a figure which shows the die map of the arrangement | sequence state of die | dye after execution of alignment of a wafer. It is a figure explaining the coordinate value update in the alignment procedure of a wafer. It is a figure explaining the rotation amount and Y direction size in the alignment procedure of a wafer. It is a figure explaining the interpolation of the focus value in the focus recipe preparation in the alignment procedure of a wafer. It is a figure explaining the error which arises in the alignment procedure of a wafer. It is a figure explaining the basic flow of the inspection procedure of a semiconductor device. It is a figure which shows the setting of inspection object die | dye. It is a figure explaining the setting of the inspection area | region inside die | dye. It is a figure explaining the test | inspection procedure of a semiconductor device. It is a figure explaining the test | inspection procedure of a semiconductor device. It is a figure which shows an example of a scan in case there is one inspection die, and an example of an inspection die. It is a figure explaining the production | generation method of the reference image in the test | inspection procedure of a semiconductor device. It is a figure explaining the adjacent die comparison method in the inspection procedure of a semiconductor device. It is a figure explaining the adjacent die comparison method in the inspection procedure of a semiconductor device. It is a figure explaining the reference | standard die comparison method in the test | inspection procedure of a semiconductor device. It is a figure explaining the reference | standard die comparison method in the test | inspection procedure of a semiconductor device. It is a figure explaining the focus mapping in the test | inspection procedure of a semiconductor device. It is a figure explaining the focus mapping in the test | inspection procedure of a semiconductor device. It is a figure explaining the focus mapping in the test | inspection procedure of a semiconductor device. It is a figure explaining the focus mapping in the test | inspection procedure of a semiconductor device. It is a figure explaining the focus mapping in the test | inspection procedure of a semiconductor device. It is a figure which shows embodiment which connected the electron beam apparatus which concerns on this invention to the semiconductor manufacturing line.

Claims (11)

In an electron beam apparatus having a mapping projection type electron optical system for inspecting the surface of a sample,
An electron gun that emits an electron beam;
A primary electron optical system for directing and irradiating the emitted electron beam on a sample;
A detector for detecting electrons;
A secondary electron optical system that guides an electron beam emitted from the sample by irradiation of the electron beam and obtained information on the surface of the sample to a detector;
At least one of the primary electron optical system and the secondary electron optical system includes a multipole lens. 2. The electron beam apparatus according to claim 1, wherein the multipole lens is disposed between the magnifying lens of the secondary electron optical system and the beam separating means for separating the primary electron beam and the secondary electron beam. An electron beam device. 3. The electron beam apparatus according to claim 2, wherein the objective lens closest to the sample surface is an electromagnetic lens having a magnetic gap provided on the sample side, and axial chromatic aberration caused by the electromagnetic lens is applied to the secondary electron optical system. An electron beam apparatus characterized in that correction can be made with a arranged multipole lens. 2. The electron beam apparatus according to claim 1, wherein the multipole lens is disposed between a reduction lens of the primary electron optical system and beam separating means for separating the primary electron beam and the secondary electron beam. An electron beam device. The electron beam apparatus according to claim 4, wherein
The primary electron optical system includes an axisymmetric lens,
The axisymmetric lens is set so that the off-axis aberration at the field edge is equal to or less than a predetermined value set in advance, and is configured by an electromagnetic lens, and the Bohr diameter of the lens is 50 times or more the maximum diameter of the field of view. An electron beam apparatus characterized by that. The electron beam apparatus according to claim 1,
The primary electron optical system includes means for converting an electron beam from an electron gun into a multi-electron beam,
The detector comprises an electron beam apparatus comprising a detector for individually detecting a multi-electron beam, which is a secondary electron beam emitted from an irradiation point on the multi-electron beam sample. 2. The electron beam apparatus according to claim 1, wherein the multipole lens is a four-stage quadrupole lens. In a defect inspection system for inspecting defects on the surface of a sample,
An electron beam apparatus having the mapping projection type electron optical system according to claim 1;
Image acquisition means for generating an image of the sample surface based on information on the sample surface included in the electrons detected by the detector of the electron beam apparatus;
A defect inspection system comprising: defect evaluation means for inspecting the presence or absence of defects on the sample surface by comparing the acquired image with a reference image. 8. The defect inspection system of claim 7, further comprising:
A sample transport system for transporting the sample;
A sample mounting unit for mounting a sample;
An XY stage for two-dimensionally moving the sample mounting unit;
A main chamber for storing the sample mounting unit and the XY stage and maintaining a vacuum state;
A defect inspection system comprising a load lock chamber that is located between the main chamber and the sample transport system and maintains a vacuum state of the main chamber when the sample is moved from the sample transport system to the main chamber. . The defect inspection system according to claim 9, wherein the sample transport system includes an electrostatic chuck having a function of preventing adhesion of particles to the sample. 8. The defect inspection system according to claim 7, wherein the XY stage includes an air bearing having a differential exhaust mechanism in at least one axial direction thereof.

Images