JP2011159636A - Method and device for inspecting pattern defects - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve a problem wherein a large charged potential may generate large contrast disorder on electron image formation since a charged distribution with a large potential may be generated on the whole wafer in a certain process of semiconductor manufacturing processes. <P>SOLUTION: In an inspection method for pattern defects, when the wafer is conveyed from a reserve chamber 202 to a vacuum chamber 201, the charge of the wafer generated in advance is eliminated by using a precharge removing device 1201, or is decreased to a potential which can be controlled with precharging performed at the time of inspection. It is preferable that an installing position of the precharge removing device 1201 is at a vacuum chamber 201 side of an opening between the reserve chamber 202 and the vacuum chamber 201. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a method and apparatus for inspecting an electrical defect of a fine circuit formed on a semiconductor wafer.

  In a semiconductor device manufacturing process, as a method for detecting a defect in a circuit pattern formed on a wafer by image comparison inspection, a so-called SEM pattern comparison inspection method that scans a point-shaped electron beam is, for example, [Patent Document 1]. The SEM type inspection device has a feature that it has higher resolution than an optical inspection device and can detect electrical defects. However, when a sample is inspected with an SEM type inspection apparatus, it is a technique for obtaining an image by narrowing the electron beam into a dot shape and scanning the sample surface two-dimensionally. There is a fundamental obstacle to speeding up.

  In addition, as an electron beam inspection method for increasing the speed, for example, in [Patent Document 2], a mapping is performed in which a semiconductor wafer is irradiated with a rectangular electron beam and the generated reflected electrons and secondary electrons are imaged by an electron lens. A mold inspection device is described. Since the mapping type inspection apparatus can irradiate an electron beam with a larger current than the SEM method at a time and can acquire images in a lump, it can be expected that an image can be formed at a higher speed than the SEM method.

  However, in the mapping type inspection apparatus, there is a problem that the distribution angle of secondary electron emission is wide when forming an image. Since the secondary electron emission angle distribution follows the cosine law, most secondary electrons are emitted at a large angle with respect to the wafer normal direction. When all these secondary electrons are taken into the objective lens and imaged, sufficient spatial resolution cannot be obtained due to the aberration of the objective lens. In order to obtain a sufficient spatial resolution of the 100 nm level, it is necessary to limit the secondary electrons emitted within a slight opening angle (for example, 0.1 rad) with respect to the axial direction of the lens to form an image. Therefore, even if a large current is irradiated with an electron beam as an area beam for batch image formation, the ratio of secondary electrons that can actually contribute to image formation is low, so the necessary S / N ratio of the image is secured. Is difficult. Even when reflected electrons are used, only an emission amount that is two orders of magnitude smaller than the irradiation beam current can be obtained, and it is difficult to achieve both high defect detection sensitivity and high speed in a normal mapping type inspection apparatus.

  On the other hand, as a technique for achieving both high-sensitivity defect detection and high-speed detection, [Patent Document 3] describes electrons (mirror electrons or mirror-reflected electrons and the following) that are pulled back before colliding with a sample by a reverse electric field directly above the wafer. A mapping type wafer inspection apparatus using the image forming electrons as imaging electrons.

  The mirror electron imaging type inspection apparatus has two main differences from the conventional mapping type inspection apparatus that forms images of secondary electrons and reflected electrons. First, since the mirror electrons from the sample do not have a wide angular distribution like secondary electrons and are emitted almost directly above the sample surface, almost all of them can be taken into the imaging lens system. It is to increase the amount. Second, in the region where the incident electrons are mirror-reflected immediately above the sample, the kinetic energy of the electrons is extremely small and the trajectory is changed even by slight surface fluctuations, so the difference in image contrast between the defective part and the normal part is large. It is to become. Compared to secondary electron and backscattered electron imaging inspection equipment that acquires images with high resolution and detects slight image differences, the difference in image contrast between the defective part and the normal part increases, so the burden of image processing is reduced. Is meant to be. In addition to these features, in the mirror electron imaging type inspection apparatus, most of the irradiation electrons are reflected directly on the wafer, and therefore basically do not enter the wafer. Since there is an energy distribution in the electron beam, there are electrons with a slight high energy, which enter through the potential barrier, which is only a few eV. That is, in the SEM type inspection apparatus, the secondary electron imaging type inspection apparatus, and the like, it is possible to inspect even a sample that is likely to be damaged by an electron beam.

  In the mirror electron imaging type inspection apparatus, a change in potential formed by surface irregularities is sensitively detected. However, as with the SEM type inspection apparatus, it is possible to provide sensitivity to electrical defects formed on the wafer. . For example, if there is a defect in electrical continuity, the location is electrically insulated, so that the potential on the surface can be made different from the normal portion that is conducting by charging, and the potential abnormality is mirror electrons. It can be detected using imaging. However, in the mirror electron imaging type inspection apparatus, since the irradiation electron beam is almost driven back by the reverse electric field immediately before the wafer, charging of the sample cannot be controlled by the irradiation electron beam. Therefore, in order to obtain a stable inspection image, it is necessary to control (preliminary charging) the charged state of the sample surface before irradiation with the primary electron beam. Pre-charging is performed by irradiating the sample to be inspected with ultraviolet light or an electron beam having energy sufficient to generate secondary electrons, or applying a predetermined potential to the sample surface. It is performed by the method of.

  [Patent Document 4] describes a preliminary charging device that charges a wafer in advance before inspection. The charged potential on the wafer surface due to the preliminary charging varies depending on the type of insulator and the circuit pattern, but decreases gradually with a certain time constant because the charge escapes little by little. These time constants are sufficiently longer than the time for acquiring the mirror electronic image, but are insufficient compared to the inspection time for the entire wafer, and pre-charging during inspection is required.

  In order to perform electrical defect detection, in order to control the charged potential of the wafer by preliminary irradiation with an electron beam, a control electrode is provided immediately above the wafer. According to [Patent Document 4], this control electrode is configured by using a grid-like electrode so as to transmit an irradiation electron beam and to apply a potential directly above the wafer. The principle of controlling the charging potential by the grid electrode will be described below. At the time of preliminary irradiation, the irradiation energy of the electron beam is set to such a value that the secondary electron emission efficiency is 1 or more. The insulating film material for general semiconductor devices is about 500V. The surface of the insulating film formed on the wafer by electron beam irradiation is positively charged because the secondary electron emission efficiency is higher than 1. When a positive potential is applied to the control electrode relative to the wafer surface potential, the generated secondary electrons are attracted to the control electrode, so that the wafer surface is positively charged. When the charged potential on the wafer surface becomes equal to the voltage on the control electrode, there is no potential gradient between the control electrode and the wafer surface, and the generated secondary electrons begin to return to the wafer surface. Then, the positive charge on the wafer surface is relaxed, the potential gradient with respect to the control grid is restored again, and the secondary electrons are attracted toward the control electrode again. Eventually, the charged potential on the surface of the wafer is balanced at a potential substantially equal to the potential of the control electrode.

  Conversely, when a negative potential is applied to the control electrode relative to the wafer surface potential, the generated secondary electrons are pushed back from the control electrode and return to the wafer surface, so that the effective secondary electron emission efficiency is less than 1. Become. Eventually, the wafer surface is negatively charged until there is no potential gradient between the control electrode and the wafer surface. Thus, the charged potential on the wafer surface can be controlled by the control electrode.

JP 05-258703 A JP-A-7-249393 JP-A-11-108864 JP 2004-14485

  In order to achieve further increase in inspection speed and defect detection accuracy in wafer pattern inspection, the prior art has the following problems.

  In the case of the mirror electron imaging type inspection apparatus, the electrical potential in the wafer due to preliminary irradiation needs to have a uniform charging potential as a whole of the preliminary irradiation region. The reason for this is that in the mirror electron imaging inspection apparatus, since the reflected electron is reflected on a potential potential surface on the wafer surface to form an image, the potential to reflect the irradiated electron beam when the charged potential of the wafer slightly changes. The height of the surface from the wafer surface varies. As a result, the imaging conditions vary, and the contrast of the mirror electron image varies. According to the results of our experiments, the variation in allowable charging potential is about 0.5 V or less, and such a uniform charging potential distribution cannot be achieved simply by placing a grid electrode and irradiating an electron beam. I know that.

  In addition, although there is an example in which a preliminary charging device is applied to a conventional SEM type inspection apparatus, in the SEM type inspection apparatus, secondary electrons generated when irradiated electrons enter the wafer during inspection recharge the irradiated region. In addition, since the secondary electron detection efficiency does not change with a sample potential fluctuation of about several volts, high-precision uniformity by preliminary charging is hardly required. Here, in the case of the mirror electron imaging type inspection apparatus, since the irradiation electron beam is almost driven back by the reverse electric field immediately before the wafer, charging of the sample cannot be controlled by the irradiation electron beam.

  In addition, in the case of the mirror electron imaging method, the required charging uniformity of the sample surface is very severe, about ± 0.5V, and the preliminary charging technology used in the conventional SEM method has sufficient uniformity. Could not get. When the conventional pre-charging technique is applied to a mirror electron imaging type device, it is considered that the following problems occur specifically.

  In the charging control device, a grid electrode is used as a control electrode, and there is a problem that a two-dimensional non-uniform distribution occurs in the charging potential. When charging is controlled by irradiating ultraviolet rays or electron beams through grid-like electrodes, the portion corresponding to the grid on the wafer is not irradiated, so that sufficient charge is not supplied and the desired charging potential is reached. The part which is not formed will be formed. The charge control by the preliminary irradiation can be executed at any timing before or during the inspection of the sample to be inspected. Needless to say, such non-uniform charging has an adverse effect on the inspection accuracy.

  In the case where preliminary irradiation is performed at the time of inspection using a preliminary charging device or the like, the non-uniformity of charging increases particularly in a direction perpendicular to the moving direction of the wafer. The wafer moves relative to the preliminary irradiation apparatus, and there are many cases where irradiation regions partially overlap in the moving direction of the wafer. Therefore, the charged state is averaged in the direction along the moving direction, and the non-uniformity is alleviated to some extent. However, since the averaging is not performed for the non-uniformity in the direction perpendicular to the moving direction, the non-uniformity of the charging potential remains. As described above, the abnormal contrast in the mirror electron image caused by the nonuniformity of the charged potential causes a problem that a non-defect is counted as a defect in an actual inspection, and a correct inspection cannot be performed.

  A means for always uniformly charging a desired area including an inspection area of a sample before acquiring an inspection image by mirror electrons is provided.

  Specifically, for example, a preliminary charging device is provided in the mirror electron imaging type inspection device, and an electrode having a slit-like opening is provided as an electrode for controlling the charging potential of the preliminary charging device, not a grid-like electrode, By setting the longitudinal direction of the opening in a direction perpendicular to the moving direction of the stage, averaging of the charged potential of the sample was achieved by moving the stage. In addition, the potential gradient at the boundary of the preliminary charging region is made gentle by reducing the irradiation intensity as it approaches the boundary of the preliminary irradiation region, and can be sufficiently uniformed by new preliminary charging.

  According to the present invention, since a uniform charging potential can be always formed on a wafer, defects in a semiconductor pattern can be detected at high speed and without error.

The figure explaining the basic composition of the mirror electronic imaging type inspection device by the present invention. The figure explaining inspection operation | movement of a mirror electronic imaging type inspection apparatus. The figure explaining the structure of a preliminary charging device. The figure explaining the Example of the extraction electrode of a preliminary charging device. The figure explaining the Example of the extraction electrode and beam shaping slit of a preliminary charging device. The figure explaining the Example of a beam shaping slit. The figure explaining the Example of a beam shaping slit. The figure explaining the Example of a preliminary charging device. The figure explaining the Example of a preliminary charging device. The figure explaining the Example of a preliminary charging device. The figure explaining the Example of a preliminary charging device. The figure explaining the Example of the mirror electronic imaging type inspection apparatus using a preliminary | backup decharging apparatus. The figure explaining the Example of a preliminary | backup charge removal apparatus. The figure explaining the Example of a preliminary | backup charge removal apparatus. The figure explaining the example of test | inspection operation | movement of the mirror electronic imaging type | mold test | inspection apparatus using a preliminary | backup charge removal apparatus. The figure explaining the subject which arises by preliminary charging. The figure explaining the Example of a beam shaping slit.

  The configuration of an embodiment of the present invention will be described in detail below with reference to the drawings.

  FIG. 1 shows an outline of a mirror electron imaging type inspection apparatus equipped with a preliminary charging device. However, a vacuum exhaust pump, its control device, exhaust system piping, etc. are omitted in this figure.

  First, the main elements of the electron optical system of this apparatus will be described. The irradiation electron beam 901 emitted from the electron gun 101 is deflected by the E × B deflector 103 while being converged by the condenser lens 102 to form a crossover 902, and then becomes a substantially parallel bundle on the sample wafer 104. Is irradiated. In the drawing, the condenser lens 102 is illustrated as one, but a system in which a plurality of lenses are combined to further optimize the optical conditions may be used. In this embodiment, a Zr / O / W Schottky electron source is used for the electron gun 101. An electron gun using a Zr / O / W Schottky electron source can stably supply a high-current beam (for example, 1.5 μA) and a uniform electron beam with an energy width of 1.5 eV or less. Suitable for the intended device. However, in the present invention, not only a Zr / O / W type electron source but also an electron source having high luminance and a small electron source size is useful. For example, an electron source using carbon nanotubes is also used. be able to. Voltage currents necessary for operation, such as the extraction voltage to the electron gun 101, the acceleration voltage to the extracted electron beam, and the heating current of the electron source filament, are supplied and controlled by the electron gun control device 105.

  The installation position of the E × B deflector 103 is set in the vicinity of the imaging surface 904 of the imaging electron beam 903. At this time, aberration is generated by the E × B deflector 103 with respect to the irradiation electron beam 901. When this aberration needs to be corrected, another aberration correcting E × B deflector 106 is disposed between the irradiation system condenser lens 102 and the E × B deflector 103.

The irradiation electron beam 901 deflected along the axis perpendicular to the wafer 104 by the E × B deflector 103 is formed into a planar electron beam incident in the direction perpendicular to the surface of the sample wafer 104 by the objective lens 107. The Since a fine crossover is formed by the irradiation condenser lens 102 on the focal plane of the objective lens 107, an electron beam with good parallelism can be irradiated onto the sample wafer 104. A region on the sample wafer 104 irradiated with the irradiation electron beam 901 has a large area such as 2500 μm ² or 10,000 μm ² .

  The sample wafer 104 mounted on the wafer stage 108 is applied with a negative potential substantially equal to or slightly lower than the acceleration voltage of the electron beam (large absolute value). The irradiation electron beam 901 is decelerated and reflected in front of the wafer 104 with almost no collision with the wafer 104 by this negative potential, and is pulled back upward as mirror electrons. Supply and control of the voltage applied to the wafer 104 is performed by the wafer voltage controller 109. In order to reflect the irradiation electrons in the very vicinity of the wafer, it is necessary to adjust the difference from the accelerating voltage of the irradiation electron beam 901 with high accuracy, and interlock control with the electron gun control device 105 is performed.

  Electrons flying from the wafer side reflect information on electrical defects in the circuit pattern on the wafer 104, and are taken into the apparatus as an image for defect determination by image formation using an electron imaging optical system. The mirror electrons are converged by the objective lens 107, and the E × B deflector 103 is controlled so as not to deflect the electron beam traveling from below. 110 and the projection lens 111 enlarges and projects the image on the image detection unit 112. In this figure, the projection lens 111 is depicted as one, but it may be composed of a plurality of electronic lenses for high magnification and image distortion correction. The image detection unit 112 converts the image into an electrical signal and sends a local distribution of charged potential on the surface of the wafer 104, that is, a defect image to the image processing unit 112.

  In addition to the electron optical system described above, an auxiliary electron optical element such as an aligner for correctly passing an electron beam through each electron optical element and a stigma for correcting image distortion are appropriately installed in the electron optical system. It is omitted in the figure. The electron optical system controller 113 controls the electron optical system including these auxiliary electron optical elements.

  Next, the image detection unit 112 will be described. For image detection, a fluorescent screen 112a for converting a mirror electronic image into an optical image and an optical image detection device 112b are optically coupled by an optical image transmission system 112c. In this embodiment, an optical fiber bundle is used as the optical image transmission system 112c. The optical fiber bundle is a bundle of thin optical fibers equal to the number of pixels, and can efficiently transmit an optical image. Further, when a fluorescent image having a sufficient amount of light can be obtained, the optical transmission efficiency may be lowered. An optical lens is used instead of the optical fiber bundle, and the optical image on the fluorescent plate 112a is optically detected by the optical lens. You may make it image-form on the light-receiving surface of 112b. In this case, the degree of freedom of the optical transmission system is increased, and image processing such as optical image enlargement and distortion correction can be performed more easily. Further, an amplification device may be inserted into the optical image transmission system so that a sufficient amount of optical image can be transmitted to the optical image detection device 112b. The optical image detection device 112b converts the optical image formed on the light receiving surface into an electrical image signal and outputs it. As the optical image detection device 112b, a CCD, an MCP (microchannel plate), a photodiode, or the like can be used. Alternatively, a TDI sensor using a time accumulation type CCD may be used.

  The wafer stage 108 must be driven in synchronization with the image capturing timing of the image detection unit 112. In particular, when a TDI sensor is used as the optical image detection device 112b, the image resolution is significantly reduced unless the transfer of the image signal between the pixels and the movement of the wafer stage 108 are correctly synchronized. Therefore, the wafer stage is provided with a position detector 114 for detecting the position of the wafer stage, and the stage control is controlled by the stage control device 115. At the same time, the image capture timing is determined based on the stage position information from the position detector 114. It transmits to the detection part 112.

  The image processing unit 116 includes an image signal storage unit 116a and a defect determination unit 116b. The image storage unit 116a acquires electro-optical conditions, image data, and stage position data from the electro-optical system control device 113, the image detection unit 112, and the stage control device 115, respectively, and acquires the image data on the coordinate system on the sample wafer. Remember to relate to. The defect determination unit 116b uses image data with coordinates on the wafer, compares with a preset value, compares with an adjacent pattern image, or compares with an image of the same pattern portion on an adjacent die, etc. The defect is determined by various defect determination methods. The coordinates of the defect and the signal intensity of the corresponding pixel are transferred and stored in the inspection apparatus control unit 117. These defect determination methods are set by the user, or the inspection apparatus control unit 117 selects a method associated with the type of wafer in advance.

  The operating conditions of each part of the apparatus are input / output from the inspection apparatus control unit 117. Various conditions such as an acceleration voltage at the time of generating an electron beam, an electron beam deflection width / deflection speed, a wafer stage moving speed, an image signal capturing timing from an image detection element, and the like are input to the inspection apparatus control unit 117 in advance. It controls the overall control system of the elements and provides an interface with the user. The inspection device control unit 117 may be composed of a plurality of computers that share a role and are connected by a communication line. In addition, an input / output device 118 with a monitor is installed.

  In the mirror electron imaging type inspection apparatus of the present embodiment, since the electron beam hardly collides with the wafer 104, the sample wafer is not sufficiently charged. However, in order to detect an electrical defect, it is necessary to perform sufficient charging to cause a difference from the normal part. Therefore, the preliminary charging device 119 is provided.

  The preliminary charging devices 119a and 119b are both controlled by the preliminary charging control device 120. In order to prevent the charged potential on the wafer formed by the preliminary charging devices 119a and 119b from disturbing the state in which the irradiated electron beam is reflected near the wafer surface pole, the preliminary charging control device 120 is connected to the wafer voltage control device 109 and the electron. Interlocking control with the gun control device 105 is performed. Details of these operations will be described with reference to FIG. FIG. 2 does not show a vacuum exhaust system, an image processing system, a stage control system, an electron optical system, a control system, and the like for maintaining the apparatus in a vacuum. FIG. 2 is a top view of the mirror electron imaging type inspection apparatus. In FIG. 2, the details of the electron optical system column are omitted, and only the position of the objective lens 107 for mirror electron imaging is shown. The wafer stage 108 accommodated in the vacuum chamber 201 moves in the Y direction little by little while reciprocating in the X direction as indicated by arrows in the drawing so that the entire wafer can be inspected. The preliminary charging devices 119a and 119b are arranged one by one on both sides of the objective lens 107. This is to efficiently pre-charge the wafer reciprocating with respect to the column immediately before the inspection, and to charge the wafer before the inspection and to remove the charge after the inspection. The control system and electrical wiring of the preliminary charging devices 119a and 119b are omitted in the drawing.

  First, the wafer 104 to be inspected is set in the wafer insertion port 203 by putting it in a wafer cassette or the like. After that, it is transferred by the transfer robot 204 to the spare chamber 202 opened to the atmosphere. Next, after the inlet of the preliminary chamber 202 is closed, the preliminary chamber 202 is evacuated by a vacuum pump, and the wafer 104 is loaded without breaking the vacuum of the sample chamber 201 that is always kept in a vacuum. At the time of carrying in the wafer, the wafer stage 108 moves to the HP position in the drawing and receives the wafer. After the wafer is transferred to the sample chamber 201, the entrance to the preliminary chamber 202 is closed.

  When a TDI sensor is used for the optical image detection device 112b, the integration direction of the pixel signal coincides with the continuous motion direction of the wafer stage 108. In addition, the arrangement of the preliminary charging device when performing the preliminary charging during the inspection needs to be along the continuous movement direction of the wafer stage. In wafer inspection, the continuous movement direction of the stage may be rotated 90 degrees depending on the pattern arrangement of the wafer to be inspected. That is, when the patterns to be inspected are densely arranged in the X direction in the figure and coarsely arranged in the Y direction, the stage movement as indicated by the arrow A in the figure may be sufficient, but the pattern arrangement is reversed. It is more desirable to move this direction 90 degrees. However, the mounting angle of the TDI sensor is usually constant, that is, in the case of FIG. 2, the integration direction of the pixel signal is the X direction, and the arrangement of the preliminary charging device 119 is also along the X direction. Thus, it cannot rotate 90 degrees in the direction of movement of the inspection. Accordingly, when the transfer robot 204 transfers the wafer set at the wafer insertion port 203 to the spare chamber 202, the transfer robot 204 rotates the wafer according to the inspection movement direction specified by the user, and places the wafer on the wafer stage of the spare chamber 202. It has a mechanism to do. Alternatively, in addition to this mechanism, two types of wafer insertion ports 203 may be provided and divided into the case of performing the inspection movement indicated by the arrow A and the case of performing the movement rotated 90 degrees. Alternatively, the transfer robot 204 always transfers the wafer to the preliminary chamber 202 with the same wafer orientation, but after the transfer, the wafer stage of the preliminary chamber 202 rotates in accordance with the inspection movement designated by the user, and the orientation of the wafer is set. The wafer stage may be transported to the vacuum chamber 201 in a manner consistent with the continuous movement direction of the wafer stage. Alternatively, a wafer rotation mechanism may be added to the wafer stage 108 itself so that the relationship set in advance by the user in the vacuum chamber 201 with respect to the wafer installation direction and the wafer stage continuous movement direction may be satisfied.

  An example in which the inspection operation after the wafer installation is completed is performed from the upper right of the wafer in FIG. First, the wafer stage 108 moves so that the first inspection site comes to the left side of the preliminary charging device 119a. When the inspection is started, the wafer stage 108 moves to the right, and the preliminary charging device 119a performs preliminary irradiation under preset conditions to sequentially charge the wafers. When the wafer passes through the objective lens 107, a mirror electron image is acquired and a defect is detected from the inspection image. Thereafter, when passing under the preliminary charging device 119b, recharging is performed if necessary. The recharging includes static elimination when the charging potential is 0V.

  When the inspection area passes through the preliminary charging device 119b, the inspection area is slightly moved upward in the figure by an amount set in advance in the Y direction, and the X direction is reversed and moved. Since the moving direction of the wafer is reversed, in this scanning, the preliminary charging device 119b is charged to a charging potential required when acquiring the mirror electron image, and the preliminary charging device 119a performs recharging. The preliminary charging control device 120 switches the conditions of the preliminary charging devices 119a and 119b in accordance with the switching of the wafer scanning direction according to a command from the inspection device control unit 117. When the wafer to be inspected is a wafer that can easily maintain a charged state, the preliminary charging device 119 may be intermittently operated during the inspection. The area charged by the preliminary charging device 119 is sufficiently wider than the inspection visual field, and a wide area can be charged only by scanning the wafer once in the X direction. Therefore, the operation of the preliminary charging device 119 can be stopped in the next and subsequent scans. The inspection device control unit 117 determines the ON / OFF timing of the preliminary charging device 119 according to the setting of the user, and instructs the preliminary charging control device 120.

  FIG. 3 shows details of the preliminary charging device 119. FIG. 3a shows a cross section of the precharging device. This is a case where the irradiation beam for performing preliminary charging is an electron beam. As the electron beam source, a planar electron source 301 typified by an electron source using carbon nanotubes (CNT) is used. The electron source 301 is installed in the vacuum chamber 201 through an insulated table 302, and a cable for applying an electron beam acceleration voltage Va is connected from outside the vacuum via an introduction terminal 303. The grid-like extraction electrode 304 is installed insulated from the vacuum chamber 201, and a cable for applying the extraction voltage Ve is led out of the vacuum via the introduction terminal 303. In this figure, the electron source 301 and the extraction electrode 304 are independently attached to the vacuum chamber 201, but those manufactured as an integral structure may be used.

  The charge control electrode 305 for controlling the charging of the wafer 104 is installed insulated from the vacuum chamber so as to be as close as possible to the wafer 104. In the case of a planar electron source such as a CNT electron source, the parallelism of the beam is good. Therefore, when a grid electrode is used, the portion that is shaded by the grid and does not transmit the electron beam is shaded to reach the wafer. As a result, a lattice-like nonuniformity occurs in the irradiation beam intensity. Since the wafer being inspected is moving in the X direction, the non-uniform distribution in the direction along the X direction is averaged, but the non-uniformity in the direction perpendicular to the moving direction (Y direction) is not averaged. In mirror electron imaging where uniformity of the charged potential is required, such non-uniformity of the irradiation electron beam intensity causes an abnormal contrast of the image and causes false information. FIG. 3 b shows the structure of the charge control electrode 305. The charge control electrode 305 has a width Wc in the direction along the stage movement direction (X direction) at the time of inspection on the conductive plate, and a direction (Y direction) perpendicular to the stage movement direction (X direction) at the time of inspection. An opening having a length Lc in the direction was provided. The length Lc was longer than the width Wc. A cable for applying the applied voltage Vc of the charge control electrode 305 is led out of the vacuum via the introduction terminal 303. The voltage applied to each electrode is supplied from the preliminary charging control device 120 in accordance with user settings.

  Further, in this embodiment, a beam shaping slit 306 is provided between the extraction electrode 304 and the charge control electrode 305. FIG. 3 c shows the structure of the beam shaping slit 306. The beam shaping slit 306 has a structure in which an opening having a width Ws and a length Ls is provided on a conductive plate. However, Ws <Wc, Ls <with respect to the width Wc and the length Lc of the charge control electrode 305. The relationship of Lc is maintained. The beam shaping slit 306 is installed to prevent electrons from the electron source 301 from colliding with the charging control electrode 305. By installing the beam shaping slit 306, it is possible to prevent a risk that secondary electrons generated when the charge control electrode 305 is exposed to the electron beam disturb the charged potential formed on the wafer 104. Furthermore, if the beam forming slit 306 is provided with a heating mechanism for preventing contamination by electron beam irradiation, abnormal deflection of the electron beam due to charging of the adsorbed material can be prevented, and uniform irradiation to the wafer is maintained. be able to.

  According to the present embodiment, since uniform electron beam irradiation can always be performed on the wafer in the preliminary charging operation, a uniform charging potential can be formed, and the accuracy of wafer inspection can be improved.

  In the present embodiment, an example will be described in which the extraction electrode in the preliminary charging device 119 is configured with a structure other than the grid electrode normally used in the apparatus configuration as shown in FIG. FIG. 4 shows the structure of the extraction electrode. It has a blind structure in which ultrafine conductive wires 402 are arranged in parallel on an opening 401 through which an electron beam passes. The direction in which the wire 402 is stretched is a direction (Y direction) perpendicular to the movement direction of the wafer stage during inspection. By mounting this embodiment on a pre-charging device as shown in FIG. 1 or FIG. 3A, a uniform electron beam from the electron beam extraction stage to the direction perpendicular to the wafer stage movement direction at the time of inspection. Irradiation can be performed on the wafer. Therefore, a charged potential with higher uniformity can be formed, and the accuracy of wafer inspection can be further improved.

  Depending on the type of wafer to be inspected, there may be a large amount of charge that must be supplied to the wafer before the desired charged potential is reached. Since the speed of the wafer stage cannot be changed in the preliminary charging at the time of inspection, in order to supply more charge, the current from the electron source is increased or the irradiation area is expanded in the direction of movement of the wafer stage. is required. Increasing the current from the electron source tends to cause inconveniences such as shortening the life of the electron source. Therefore, in this embodiment, in the preliminary charging device as shown in FIG. 1 or FIG. 3A, the beam irradiation area is widened by using the charging control electrode 501 and the beam shaping slit 502 provided with a plurality of openings. . FIG. 5 shows the configuration of the charge control electrode 501 and the beam shaping slit 502. The pitch P of the openings of the charging control electrode 501 and the beam shaping slit 502 (the distance P from the opening start point A of the opening in the stage motion direction (X direction) to the opening start point B of the adjacent opening) is made the same, and the charging control electrode The beam forming slit 502 can be adjusted so that the electron beam passes only inside the opening 501.

  If this embodiment is applied to the apparatus shown in FIG. 1, the amount of electron beams supplied to the wafer can be increased without imposing a burden on the electron source while maintaining the uniformity of the charged potential.

  In the present embodiment, an example will be described in which the amount of electrons irradiated to the wafer is asymptotically decreased toward the outside in the preliminary charging device 119 of the device as shown in FIG. 1 or FIG.

  Conventionally, the preliminary charging device in the mirror electron imaging type inspection apparatus is disposed in the vicinity of the imaging electron optical system column. The inspection wafer moves little by little in the Y direction while reciprocating in the X direction, and the entire surface of the wafer is inspected. The area to be pre-charged is sufficient compared to the field of view for inspection in order to eliminate the influence of the boundary of the pre-irradiation area on the potential distribution in the mirror electron imaging field, that is, to form a uniform charged potential in the inspection area Must be bigger. For example, the field size of the mirror electron imaging in the inspection is 100 to 200 microns, but the size of the pre-irradiation region is about 10 mm, which is a large difference. As shown in FIG. 16, during the continuous movement of the wafer in the X direction, the precharged region is connected to the surface of the wafer in a band shape. The portion that has already been subjected to preliminary irradiation is charged to a predetermined potential, and the portion that has not been subjected to preliminary irradiation has a potential different from the potential of the portion that has already been subjected to preliminary irradiation. In such a state, since the width of the band is larger than the visual field of the inspection, the boundary of the charged region is applied to the uninspected region that must be inspected. Since the mirror electron imaging type inspection apparatus forms the potential distribution on the surface as an image, if the boundary of the precharged region where the potential gradient exists is observed as it is with the mirror electrons, it appears as a line in the image. The sharper the change in the charge distribution in this boundary region, the stronger the linear contrast appearing in the mirror electron image. For example, in the case of a wafer on which a material having a long charge charge diffusion time is formed, a steep change in potential at the boundary between regions charged by preliminary charging is maintained. Preliminary charging is performed immediately before inspecting the pattern that overlaps the boundary area, but it takes time to completely eliminate the difference in the amount of charged charge that has occurred in advance, and an image is acquired with this line remaining. Will do. In such a case, even if preliminary charging is performed immediately before, a boundary remaining as a result of preliminary charging performed in the previous inspection appears in the mirror electronic image as a contrast abnormality at the time of subsequent inspection image acquisition. Due to such an abnormal contrast in the mirror electron image due to the nonuniformity of the charged potential, a non-defect is counted as a defect in actual inspection, and there is a problem that correct inspection cannot be performed.

  Therefore, in the present embodiment, in the preliminary charging device 119 of the apparatus as shown in FIG. 1 or FIG. 3A, the opening shape of the beam forming slit 601 is as shown in FIG. A width changing portion 602 is provided. By making the opening in this shape, the amount of electrons irradiated to the wafer is asymptotically reduced toward the outside, and as shown in FIG. Thus, contrast due to a steep potential gradient can be prevented. In addition, the beam shaping slit 603 shown in FIG. 6b is provided with a plurality of openings having a shape as shown in FIG. 6a. By using these slits, the amount of charge supplied to the wafer without generating a steep potential at the boundary. Can be increased. However, in any case, the structure of the charge control electrode may be a rectangular opening as in FIG.

  In Example 4, the opening shape of the beam forming slit was made narrower toward the both ends. However, when a plurality of openings are provided as shown in FIG. Therefore, in this embodiment, in the preliminary charging device 119 of the device as shown in FIG. 1 or FIG. 3A, the lengths of the plurality of openings are changed little by little as shown in FIG. According to the present embodiment, the amount of charge supplied to the wafer can be increased without generating a steep potential at the boundary, and preliminary irradiation that does not cause a steep potential change can be realized with easier processing.

  In the embodiments so far, an extraction electrode for extracting pre-irradiated electrons has been arranged. In this embodiment, in the preliminary charging device 119 of the apparatus as shown in FIG. 1, instead of the extraction electrode, the slit 701 itself is arranged in the vicinity of the electron source as shown in FIG. The slit doubles as a lead electrode. With this arrangement, in addition to simplifying the structure, non-uniformity of the irradiation electron distribution caused by the grid of the extraction electrode can be removed from the root, and the irradiation electron dose can be made more uniform. be able to.

  In the present embodiment, a blanker 901 is disposed in the preliminary charging device 119 of the apparatus as shown in FIG. 1 between the extraction electrode combined slit in the invention of Embodiment 6 and the charge control electrode as shown in FIG. Pre-irradiation is performed during inspection, but for wafers with a sufficiently long charge retention time, it is not necessary to perform pre-irradiation continuously during inspection, and pre-irradiation is stopped during the pre-measured retention time. Also good. That is, preliminary irradiation is performed intermittently at time intervals during the inspection. At that time, if the generation of the electron beam from the electron source itself is stopped, the state of the power supply and the electron source will slightly change the next time the electron beam is generated, and the same irradiation intensity as before the stop May not be reproduced. In addition, repeated ON / OFF of the electron source may shorten the life of the electron source itself. Therefore, in this embodiment, a blanker 901 is installed as means for stopping irradiation of the electron beam to the wafer without stopping generation of electrons from the electron source. In order to stop the irradiation of the electron beam on the wafer, a preset voltage is applied to the blanker 901 to largely deflect the electron beam trajectory, thereby shielding the electron beam by the blanker 901 side wall. The irradiation timing is controlled by the preliminary charging control device 120. According to this embodiment, it is possible to irradiate the wafer with a constant electron dose intermittently without impairing the lifetime of the electron source.

  In this embodiment, in the preliminary charging device 119 of the apparatus as shown in FIG. 1 or FIG. 3A, a predetermined opening 1701 as shown in each of the above-described embodiments is provided in advance as a beam shaping slit as shown in FIG. A plurality of movable plates 1702 are used. When only the wafer 1711 having the same chip size and a large difference in the time required for standby power is inspected, it is not necessary to change the size of the opening frequently. In the case of shifting to inspection of a wafer that takes more time, an opening suitable for a new wafer is selected, and the inspection is performed by moving the opening to a desired position. During the inspection, a method of moving in the Y direction little by little while reciprocating in the X direction may be used so that the entire wafer can be inspected. However, for the movement in the Y direction, the movement of the movable plate is added to the movement of the wafer 1703 under inspection. In addition, the movable plate 1702 may be moved by a driving device. These controls are performed by the preliminary charging control device 120. By adopting this configuration, it is possible to change the size of the irradiation region simply by selecting the opening provided in the movable plate, and the inspection speed can be increased.

  The embodiments so far have assumed an electron beam as an irradiation beam for precharging. The charging of the wafer can be controlled in the same manner as the electron beam irradiation by ultraviolet irradiation. FIG. 10 shows a case where ultraviolet rays are used as an irradiation source other than an electron source as a preliminary charging device 119 of the apparatus as shown in FIG. The ultraviolet light source 1001 generates ultraviolet light having sufficient energy to excite the photoelectrons of the wafer. The ultraviolet light source 1001 is controlled by a controller 1002, and this controller performs setting of the intensity of ultraviolet light and irradiation time, ON / OFF of ultraviolet light, and the like according to instructions from the preliminary charging control device 120. Usually, since such ultraviolet rays are vacuum ultraviolet rays that do not transmit through the atmosphere, a light source is installed in the vacuum chamber. An outer cylinder 1003 is provided for shielding the ultraviolet light so that it does not irradiate unnecessary portions, and a charge control slit 1004 is disposed at the tip protruding toward the wafer side. According to this embodiment, charging can be performed using ultraviolet rays that are more stable than electron beams.

  In the ninth embodiment, ultraviolet rays are emitted from an ultraviolet light source, and the amount of ultraviolet rays that can actually contribute to charging is small, and depending on the wafer, there is a possibility that charges necessary for desired charging cannot be generated. Therefore, in this embodiment, as shown in FIG. 11, a reflecting mirror 1101 is installed in an ultraviolet light source, and the emitted ultraviolet light is condensed so as to reach the wafer. If this embodiment is applied to the apparatus as shown in FIG. 1, the wafer can be efficiently irradiated with ultraviolet light, and the wafer can be charged in a shorter time.

  In the embodiments so far, the case where the wafer is simultaneously charged during the inspection has been described mainly for the electrical defect inspection. However, in this case, since the beam irradiation time is limited by the moving speed of the wafer, setting a long beam irradiation time has a limit. Further, depending on the process in the semiconductor manufacturing process, a charge distribution having a large potential may be generated on the entire wafer. Such a large charged potential cannot be eliminated by preliminary irradiation at the same time as an inspection in which only a limited irradiation time can be obtained, and a large contrast abnormality may occur in mirror electron imaging.

  Therefore, in this embodiment, as shown in FIG. 12, when the wafer is transferred from the preliminary chamber 202 to the vacuum chamber 201, the preliminary charging device 1201 is used to erase the charge of the wafer generated in advance or at the time of inspection. The potential is reduced to a controllable potential by preliminary charging. The installation position of the preliminary decharging device 1201 is preferably on the vacuum chamber 201 side of the opening between the preliminary chamber 202 and the vacuum chamber 201. By adopting such an arrangement, it is possible to remove the charge by preliminary charging during wafer conveyance, so that there is no great hindrance to the inspection time.

  A basic configuration of the preliminary decharging device 1201 is shown in FIG. FIG. 13 is a view of the wafer inlet 1304 from the spare chamber side from the apparatus main chamber side, and the wafer 104 is placed on the moving table 1303 and moved from the back side to the front side. In order to uniformly irradiate an electron beam over the entire surface of the wafer, an electron source 1301 longer than the diameter of the wafer was used. The electron source 1301 is, for example, a carbon nanotube grown on a rectangular substrate. Although not shown, a grid-shaped extraction electrode is provided in the vicinity for extracting electrons from the carbon nanotube, and the rectangular cross section is provided. Can generate an electron beam. A charge control electrode 1302 is provided immediately above the wafer. The charge control electrode is composed of a rectangular opening and a mesh, and forms a potential of approximately 0 V directly above the wafer. There is no need to form a uniform charge distribution just before the inspection as in the pre-charging device in the embodiments so far, and it is only necessary to control the wafer potential to the extent that it can be re-formed by the pre-charging during the inspection. A mesh electrode is used as the control electrode. The details of the structure for supporting each member and the cable for supplying power to the electrodes are omitted. The voltage applied to these electron sources and electrodes is controlled and supplied by a preliminary charge removal control device 1305 controlled by the inspection device control unit 117.

  In addition to the electron source having a large rectangular size as in the above embodiment, the electron source of the preliminary decharging device 1201 can be configured using an electron source used in the preliminary charging device 119. FIG. As shown in FIG. 14, a plurality of electron sources 1401 each provided with an extraction electrode are alternately arranged in at least two rows so that irradiation regions overlap. When sufficient electron beam irradiation is required, the number of parallel electron source arrays is increased as appropriate. In this case, in addition to the electron source 1401, a plurality of vacuum ultraviolet light sources may be arranged.

  The main flow of wafer inspection when this embodiment is used is shown in FIG. Although not shown in the drawing, various conditions in each operation are transmitted from the inspection apparatus control unit to each control apparatus in advance or as appropriate. When a wafer is introduced from the wafer transfer robot while the spare chamber is open to the atmosphere, the wafer inlet of the spare chamber is sealed and evacuated. At the same time, a preliminary charge removal operation is started to prepare for charge removal during wafer transfer. When the preliminary chamber reaches a sufficient degree of vacuum, the gate valve between the preliminary chamber and the inspection apparatus main chamber is opened, and the transfer of the wafer is started. The moving speed at the time of carrying the wafer is appropriately adjusted so that sufficient charge removal is performed. When the wafer is completely transferred to the apparatus main chamber, the beam of the preliminary decharging apparatus is stopped or cut off from the wafer, the beam irradiation is finished, and the gate valve is closed at the same time. The wafer waits at the home position (HP) for the start of inspection, and then moves to the inspection start position. Thereafter, the operation of the preliminary charging device starts, and the inspection is started on the condition that a sufficiently stable irradiation beam intensity and a voltage condition according to the recipe are obtained. When the inspection of the wafer while pre-charging is completed, the pre-charging device stops the beam or is disconnected from the wafer, and the wafer is returned to the standby position (HP) again. If there is no re-inspection of the wafer, the operation is again started so that the preliminary decharging device can irradiate the wafer with the beam. At this time, the electrode conditions are set so that the potential of the wafer is 0V. If the vacuum in the spare chamber is sufficiently good, the gate valve between the apparatus main chamber and the spare chamber is opened, and the static elimination operation and transfer of the wafer are performed. After the wafer is completely moved to the spare chamber, the beam of the precharger is stopped or interrupted. The gate valve between the apparatus main chamber and the spare chamber is closed to release the spare chamber to the atmosphere, and then the wafer is taken out of the spare chamber by the transfer robot and returned to the wafer insertion port. At this stage, the wafer is charged almost uniformly at 0 V, and does not affect the next process.

  According to this embodiment, even if unintentional charges generated by various processes are formed on the wafer in advance, not only can a stable and accurate inspection be performed, but also the influence of charging on the next process can be achieved. Can be removed.

  The embodiments of the present invention have been described above. Combinations of these examples are also included in the present invention.

  101: electron gun, 102: condenser lens, 103: E × B deflector, 104: sample wafer, 105: electron gun control device, 106: E × B deflector, 107: objective lens, 108: wafer stage, 109: Wafer voltage control device, 110: intermediate lens, 111: projection lens, 112: image detection unit, 112a: fluorescent plate, 112b: optical image detection device 112c: optical image transmission system, 113: electro-optical system control device, 114: position detection 115: Stage control device 116: Image processing unit 116a: Image signal storage unit 116b Defect determination unit 117: Inspection device control unit 118: Input / output device with monitor 119a: Preliminary charging device 119b: Pre-charging device, 120: Pre-charging control device, 201: Vacuum chamber, 202: Pre-chamber, 203: Wafer inlet, 20 4: Wafer transfer robot, 301: Electron source, 302: Stand, 303: Introduction terminal, 304: Extraction electrode, 305: Charge control electrode, 306: Beam shaping slit, 401: Opening, 402: Wire, 501: Charge control electrode , 502: beam shaping slit, 601: beam shaping slit, 602: aperture width changing portion, 603: beam shaping slit, 701: beam shaping slit, 801: extraction electrode, 901: blanker, 1001: ultraviolet light source, 1002: controller, 1003: Outer cylinder, 1004: Charge control electrode, 1101: Reflector, 1201: Preliminary charge remover, 1301: Electron source, 1302: Charge control electrode, 1303: Moving table, 1304: Wafer inlet to the apparatus main chamber, 1305: preliminary decharging control device, 1401: electron source, 901: irradiated electron beam, 902 Crossover, 903: imaging electron beam, 904: the image plane, HP: home position.

Claims (5)

A first electron beam is irradiated in a planar shape onto the first region of the sample introduced from the introduction port, the first electron beam is reflected immediately before entering the sample, and the reflected electrons are imaged. In a defect inspection method using a defect inspection apparatus for acquiring an image of a circuit pattern formed on the sample by inspecting the defect existing in the circuit pattern based on the circuit pattern image,
Before inspecting the first region, a step of irradiating the second region including the first region on the sample surface with ultraviolet rays or a second electron beam to make the charge distribution in the first region uniform. Have
A defect inspection method comprising a step of irradiating a third electron beam or ultraviolet light in the vicinity of the introduction port. The defect inspection method according to claim 1,
A defect inspection method, wherein an irradiation region of the third electron beam or ultraviolet ray is longer than a diameter of the sample. The defect inspection method according to claim 1 or 2,
A defect inspection method, wherein the third electron beam irradiation is performed by a plurality of electron sources or ultraviolet light sources. An electron optical system for irradiating the first region of the sample introduced from the introduction port with an electron beam;
A sample stage for holding the sample;
Means for applying a voltage to the sample stage or the sample such that the electron beam applied to the sample is reflected without entering the sample;
Means for detecting electrons reflected from the sample side by the voltage application;
Based on the detection signal of the detection means, having a means for forming an inspection image and detecting defects in the sample,
Means for irradiating the first region with ultraviolet rays or a second electron beam before forming the inspection image;
An inspection apparatus comprising means for irradiating a third electron beam or vacuum ultraviolet light in the vicinity of the introduction port. The inspection apparatus according to claim 4,
The third electron beam irradiation means comprises a plurality of electron sources arranged in parallel.

Images