JP2008002932A - Sample imaging apparatus and sample illumination method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a sample imaging apparatus capable of acquiring the image due to different kinds of illuminations, while suppressing the effects on the throughput and capable of being constituted easily, and to provide a sample illumination method. <P>SOLUTION: The sample-imaging apparatus 1 is equipped with an imaging element 15, constituted so that the pixel signal which forms an odd number field and a pixel signal which forms an even number field are extracted alternately from a light-detecting part and an illumination light changeover device 25 for changing-over illumination light for illuminating the surface of a sample, in a synchronous relation to the change-over of the fields, wherein the pixel signals are extracted of the light-detecting part. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an imaging apparatus for imaging a sample illuminated by bright-field illumination and dark-field illumination and a method for illuminating the sample, and in particular, an appearance inspection in which a pattern formed on a sample surface such as a semiconductor wafer is inspected in a semiconductor manufacturing process or the like. The present invention relates to an imaging apparatus used in the apparatus and a sample illumination method.

  In the semiconductor manufacturing process, a predetermined pattern is repeatedly formed on the surface of a sample such as a semiconductor wafer, a semiconductor memory photomask, or a liquid crystal display panel substrate. Therefore, an appearance inspection is performed in which an optical image of this pattern is captured and a pattern defect is detected by comparing corresponding patterns that should be the same. In such a pattern comparison type appearance inspection, if there is no difference between the two patterns as a result of the comparison, it is determined that there is no defect, and if there is a difference, it is determined that any one of the patterns has a defect.

  In the following description, a semiconductor wafer appearance inspection apparatus for inspecting a defect of a pattern formed on a semiconductor wafer will be described as an example. However, the present invention is not limited to this, and can also be applied to an appearance inspection apparatus for inspecting defects of samples such as a semiconductor memory photomask and a liquid crystal display panel substrate. The scope of application of the present invention is not limited to an appearance inspection apparatus, and is generally applicable to any imaging apparatus that captures a bright field illumination image and a dark field illumination image of a sample.

  In a semiconductor wafer appearance inspection apparatus, a metal microscope is generally used to capture an optical image of a fine pattern formed on a wafer. In such a metal microscope, bright field illumination that illuminates the target surface from the vertical direction and captures the image of the specularly reflected light, and an imaging sensor that does not detect the specularly reflected light by illuminating the sample surface from an oblique direction or a vertical direction. Arranged dark field illumination is used. FIG. 1A shows a schematic configuration example of the bright field illumination optical system, and FIG. 1B shows a schematic configuration diagram of the bright field illumination optical system.

  In the example of the bright field illumination optical system shown in FIG. 1A, in order to illuminate the sample 2, a light source 21, an illumination lens 22, and a semi-transparent mirror 24 provided on the optical axis of the objective lens 14 are provided. Is provided, and the sample 2 is illuminated from directly above through the objective lens 14 by reflecting the illumination light from the light source 21 toward the objective lens 14 by the semi-transparent mirror 24. The specularly reflected light reflected from the surface of the sample 2 enters the objective lens 14 again and is projected, and the projected light passes through the semi-transparent mirror 24 and forms an image on the light receiving surface light of the image sensor 15.

  On the other hand, in the example of the bright-field illumination optical system shown in FIG. 1B, an annular shape that is provided on the optical axis of the light source 31, the illumination lens 32, and the objective lens 14 and is hollow in the vicinity of the optical axis. The perforated mirror 34 and the illumination light that has been reflected by the perforated mirror 34 and formed into a hollow light beam that travels in the direction of the sample 2 parallel to the optical axis of the objective lens 14 is further reflected to reflect the objective lens 14. Mirrors 35 and 35 that are incident on the sample 2 obliquely with respect to the optical axis of the objective lens 14 from the outside.

The image obtained by such dark field illumination has the advantage that the relative signal intensity of the scattered light reflected by the minute defect is strong because it does not capture the specularly reflected light. The obtained image has many kinds of detectable defects, and has an advantage that a large amount of information can be used when the detected defects are subsequently classified by an automatic defect classification (ADC) device or the like.
As described above, bright field illumination and dark field illumination have their respective characteristics, and a microscope apparatus that realizes both illumination by switching between bright field illumination and dark field illumination has been devised (for example, Patent Documents 1 to 3 below). 4).

JP 2006-023221 A JP 2003-149169 A JP 2004-101406 A JP 2004-101403 A

However, according to the method of switching between bright field illumination and dark field illumination as in the above-described conventional technology, in order to obtain both a bright field image and a dark field image of one sample, the illumination method is switched and the image is taken twice. The throughput of the apparatus is reduced.
In addition, it is possible to arrange a camera at the position where the specularly reflected light reflected by the sample is captured and the position where the scattered light is captured, and simultaneously capture a bright-field image and a dark-field image. In this case, a plurality of imaging devices are required, resulting in an increase in device cost.
Furthermore, if bright field illumination and dark field illumination are applied simultaneously, the exposure amount may exceed the dynamic range of the image sensor and blooming may occur, so there are restrictions on each illumination condition and each illumination condition is optimized individually There is a problem that you can not.

  In view of the above problems, an object of the present invention is to provide a sample imaging device and an illumination method that can acquire images by different types of illumination while suppressing the influence on throughput and can be easily configured.

In order to achieve the above object, in the present invention, as an image sensor for imaging a sample, an image sensor in which a pixel signal forming an odd field and a pixel signal forming an even field are alternately extracted from a light receiving unit, The illumination light for illuminating the sample surface is switched in synchronization with the switching of the field from which the pixel signal is extracted from the light receiving unit of the image sensor.
As such an image sensor, for example, a general CCD image sensor can be used.

  For example, in a CCD image sensor, when an image signal for one frame is generated, the signal charges that form the image signal of these fields by changing the time (timing) between the odd field and the even field constituting this frame. Is taken out from the light receiving element. Therefore, by changing the type of illumination in synchronism with the switching of the field from which the photoelectric signal is extracted, an optical image generated by a plurality of different types of illumination light can be obtained by generating an odd-numbered field and an even-numbered field while generating an image signal for one frame Can be generated separately.

According to the present invention, it is possible to generate images generated by a plurality of types of illumination as images of a plurality of different fields while generating an image generated by one type of illumination as a single frame image. It is possible to acquire images obtained by a plurality of different types of illumination while suppressing the influence on the throughput.
In addition, by switching between these different types of lighting, it is possible to avoid problems such as exceeding the dynamic range and blooming caused by illuminating these lights simultaneously. It becomes possible to optimize.

  For this reason, in the sample imaging device according to the first embodiment of the present invention, the pixel signal forming the odd field and the pixel signal forming the even field are alternately extracted from the light receiving unit, and the pixel signal is extracted from the light receiving unit. And an illumination light switching unit that switches illumination light that illuminates the sample surface in synchronization with field switching.

  In addition, in the sample illumination method according to the second aspect of the present invention, the imaging of the sample surface is performed using an imaging device in which pixel signals forming odd fields and pixel signals forming even fields are alternately extracted from the light receiving unit. In the apparatus, the illumination light for illuminating the sample surface is switched in synchronization with the switching of the field for extracting the pixel signal from the light receiving unit.

  According to the present invention, it is possible to acquire images by different types of illumination without causing a large increase in apparatus cost and a large effect on throughput with respect to a conventional sample imaging apparatus.

Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. FIG. 2 is an overall configuration diagram showing an outline of the sample imaging apparatus according to the first embodiment of the present invention.
The sample imaging apparatus 1 is preferably incorporated in an appearance inspection apparatus (not shown) for performing an appearance inspection of the sample 2 such as a semiconductor wafer, a semiconductor memory photomask, and a liquid crystal display panel substrate in a semiconductor manufacturing process. Used.
In such an appearance inspection apparatus, a pattern defect is detected by comparing corresponding patterns that should be the same among the images of the surface of the sample 2 imaged by the sample imaging apparatus 1. That is, if there is no difference between the two patterns as a result of the comparison between the patterns, it is determined that there is no defect, and if there is a difference, it is determined that there is a defect in one of the patterns.
A CCD image sensor 15, which is an area sensor, is provided above the stage 11, and the CCD image sensor 15 generates an image signal corresponding to the optical image of the surface of the sample 2 formed on the light receiving surface.

  The sample imaging device 1 includes an objective lens 14 for projecting and forming an optical image of the sample 2, a stage 11 movable at least in a two-dimensional plane perpendicular to the optical axis of the objective lens 14, and the upper surface of the stage 11. And a sample stage (chuck stage) 12 provided on the surface. On the sample stage 12, the sample 2 which is a semiconductor wafer or the like to be inspected is placed and fixed. The stage 11 moves under the control of the stage drive unit 13, and the stage drive unit 13 adjusts the position of the stage 11 so that the desired observation position of the sample 2 falls within the field of view of the objective lens 14.

  The sample imaging device 1 includes a bright-field illumination optical system including a first pulse laser generator 21, illumination lenses 22 and 23, and a semi-transparent mirror (beam splitter) 24, and a second pulse laser generator. 31, a dark field illumination optical system including an illumination lens 32 and 33, a perforated mirror 34, and a mirror 35, and a laser for each of the first pulse laser generator 21 and the second pulse laser generator 31. And an illumination pulse generation circuit 25 for generating a pulse signal instructing generation.

  In the bright field illumination optical system, the illumination light generated by the first pulse laser generator 21 according to the pulse signal generated by the illumination pulse generation circuit 25 is incident on the semi-transparent mirror 24 via the illumination lenses 22 and 23. . The semi-transparent mirror 24 is provided on the optical axis of the objective lens 14, and reflects the incident laser light parallel to the optical axis of the objective lens 14 and toward the objective lens 14, thereby passing the sample 2 through the objective lens 14. Is illuminated from directly above. The specularly reflected light reflected from the surface of the sample 2 is incident on the objective lens 14 again and projected, and the projected light passes through the semi-transparent mirror 24 to form a bright field image and is formed on the light receiving surface light of the CCD image sensor 15. To do.

  In the dark field illumination optical system, the illumination light generated by the second pulse laser generator 31 in accordance with the pulse signal generated by the illumination pulse generation circuit 25 is incident on the perforated mirror 34 via the illumination lenses 32 and 33. Let The perforated mirror 34 is an annular reflecting mirror that is provided on the optical axis of the objective lens 14 and is hollow in the vicinity of the optical axis. The illumination light reflected by the perforated mirror 34 is near the optical axis of the objective lens. It becomes an annular light beam that has become hollow, and proceeds in the direction of the sample 2 in parallel with the optical axis of the objective lens 14. The annular illumination light is reflected by the mirrors 35 and 35 and enters the sample 2 obliquely with respect to the optical axis of the objective lens 14 from the outside of the objective lens 14. For this reason, the specularly reflected light of the illumination light incident on the sample 2 obliquely from the outside of the objective lens 14 does not enter the objective lens 14, and the objective lens 14 captures only the scattered light reflected on the surface of the sample 2 and detects the CCD. A dark field image is formed on the light receiving surface of the image sensor 15.

In this embodiment, by arranging the bright field illumination optical system and the dark field illumination optical system as described above, both the bright field image by the bright field illumination optical system and the dark field image by the dark field illumination optical system are the same. An image is formed on the light receiving surface of the CCD image sensor 15.
Also, an illumination pulse generation circuit for generating a pulse signal instructing each of the first pulse laser generator 21 and the second pulse laser generator 31 to provide the bright field illumination and the dark field illumination, respectively, to generate a laser. It is possible to switch between bright-field illumination and dark-field illumination by alternately providing pulse signals to these pulse laser generators 21 and 31. Further, by generating these illumination lights with a pulse laser generator, it is possible to instantaneously switch between bright field illumination and dark field illumination.

  Note that the above-described configuration of the illumination system is merely illustrated for explanation, and the scope of the present invention is not limited to this configuration example. It is possible to form a bright field image by the bright field illumination optical system and a dark field image by the dark field illumination optical system on the same light receiving surface of the CCD image sensor 15, and switch between bright field illumination and dark field illumination. Any configuration can be applied to the present invention as long as it has a certain configuration.

A field shift pulse, a transfer pulse, and a reset pulse applied to the vertical transfer CCD and the horizontal transfer CCD in order to read an image signal from the CCD image sensor 15 are generated by a timing pulse generation circuit 17. Hereinafter, a read operation of a general interline transfer CCD (IT-CCD) will be described with reference to FIGS.
FIG. 3 is a configuration diagram of a part of the CCD image sensor 15 shown in FIG. As shown in the figure, the CCD image sensor 15 receives light receiving portions R11 to R24 such as photodiodes that photoelectrically convert light hitting each pixel portion of the light receiving surface, and signal charges extracted from the light receiving portions R11 to R24 through a shift gate. The vertical transfer CCDs V11 to V28 that transfer in the downward direction in the figure and the vertical transfer CCDs V11 and V21 in the last column are arranged adjacent to each other, and the signal charges transferred from these vertical transfer CCDs V11 and V21 are transferred in the left direction in the figure Horizontal transfer CCDs H11 to H22 are arranged.

In general, the IT-CCD image sensor 15 has resolution lines that are thinned out every other row among all resolution lines of all pixels of the IT-CCD image sensor 15 in response to a television signal for performing interlace scanning. Interlaced reading is performed to alternately read odd frames and even frames formed by resolution lines thinned out by the odd frames.
As will be described later, a “frame reading method” and a “field reading method” are used for interlaced reading, and both reading methods are applicable to the present invention.
In the frame readout method, odd frames are generated from signals of pixels in odd-numbered rows (hereinafter, sometimes referred to as “odd-column pixels” for explanation), and pixels in even-numbered rows (hereinafter, “ An even frame is generated from the signal of “even column pixels”.
In the field readout method, in each of the odd-numbered frame and the even-numbered frame, when the field is generated, the accumulated charges of all the light receiving portions are read and the accumulated charges of the light receiving portions of two adjacent pixels are added. At this time, the combination of columns to be added is changed between the odd field and the even field.
In the following description, the light receiving portions R11, R13, R21, and R23 shown in FIG. 3 are assumed to be light receiving portions of odd-numbered pixels (hereinafter sometimes referred to as “odd-numbered light receiving portions”), and the light receiving portions R12, R14, R24, and R24 is a light-receiving portion of an even-numbered pixel (hereinafter may be referred to as “even-numbered light-receiving portion”).

In the CCD image sensor 15, a voltage φV2 for applying a vertical transfer pulse and a shift charge pulse is applied to the vertical transfer CCDs V12, V16, V22, and V26 that take out signal charges from the odd-numbered light receiving units R11, R13, R21, and R23, respectively. Then, a voltage φV4 for applying a vertical transfer pulse and a shift charge pulse is applied to the vertical transfer CCDs V14, V18, V24, and V28 that take out signal charges from the even-numbered light receiving units R12, R14, R22, and R24, respectively.
A voltage φV1 for applying a vertical transfer pulse is applied to the CCDs V11, V15, V21 and V25, and a voltage φV3 for applying a vertical transfer pulse is applied to the CCDs V13, V17, V23 and V27.
Further, a voltage φH1 for applying a horizontal transfer pulse is applied to the horizontal transfer CCDs H11 and H21, and a voltage φH2 for applying a horizontal transfer pulse is applied to the horizontal transfers CCDH12 and H22.

  4A to 4F are time charts showing an interlace reading operation in the CCD image sensor 15 shown in FIG. 3, and FIGS. 5A to 5F are time charts. It is explanatory drawing of the read-out operation | movement at the time of reading the signal charge of CCD image sensor 15 shown in FIG. 3 by a frame read-out system.

  4A shows the timing of applying the shift charge pulse to the vertical transfer CCDs V12, V16, V22, and V26 for extracting signal charges from the odd-numbered light receiving units R11, R13, R21, and R23, respectively. During a period of “1”, a high potential φV2 is applied, and the signal charges accumulated in the odd-numbered light-receiving portions R11, R13, R21, and R23 are shifted to the vertical transfer CCDs V12, V16, V22, and V26, respectively, and odd fields Is shifted into the vertical transfer CCD.

  FIG. 4B shows a period in which the vertical transfer pulses and the horizontal transfer pulses are applied to the vertical transfer CCDs V11 to V28 and the horizontal transfer CCDs H11 to H22, respectively. In the period “1” shown in FIG. The signal charge is transferred and taken out as an output signal.

  FIG. 4C shows the timing of applying the shift charge pulse to the vertical transfer CCDs V14, V18, V24, and V28 for extracting signal charges from the even-numbered light receiving units R12, R14, R22, and R24, respectively. During this period, the high potential φV4 is applied, and the signal charges accumulated in the even-numbered light receiving units R12, R14, R22, and R24 are shifted to the vertical transfer CCDs V14, V18, V24, and V28, respectively. The signal charge is shifted into the vertical transfer CCD.

  FIG. 4D shows the timing at which the reset pulse is applied, and the signal charges accumulated in each of the light receiving portions R11 to R24 during the period “1” shown in the figure are ejected toward the substrate and erased.

  At time t1 shown in FIG. 4, when a shift charge pulse is applied to the vertical transfer CCDs V12, V16, V22, and V26 by applying a high potential φV2, signals accumulated in the odd-numbered light receiving units R11, R13, R21, and R23 The charges are shifted to the vertical transfer CCDs V12 and V11, V16 and V15, V22 and V21, and V26 and V25, respectively (see FIG. 5A).

  Thereafter, when a vertical transfer pulse (four-phase drive waveform) is applied to each of the CCDVs 11 to 28 in the period t2 to t4, the signal charge accumulated in each vertical transfer CCD is transferred in the vertical direction pixel by pixel (FIG. 5). (See (A)). As a result, the signal charges transferred to the horizontal transfer CCDs H11 and H21 are transferred in the horizontal direction by applying a horizontal transfer pulse (two-phase drive waveform) to each of the CCDs H11 to H21 and sequentially taken out as output signals (FIG. 5 (C) and FIG. 5 (D)).

When the horizontal transfer CCDs H11 to H22 become empty, a vertical transfer pulse is again applied to the CCDs V11 to 28, and the signal charges accumulated in the vertical transfer CCDs are transferred in the vertical direction pixel by pixel ((E) in FIG. 5). reference).
By repeating the above operation in the period t2 to t4, all the signal charges taken out from the odd-numbered light receiving units R11, R13, R21, and R23 at time t1 are taken out, and the signal charges forming the odd field are converted into the CCD image sensor 15 Taken from.

At time t4, when a high potential φV4 is applied to the vertical transfer CCDs V14, V18, V24 and V28 and a shift charge pulse is applied, the signal charges accumulated in the even-numbered light receiving units R12, R14, R22 and R24 are vertically transferred. Shifted to CCDV14 and V13, V18 and V17, V24 and V23, and V28 and V27, respectively (see FIG. 5F).
Thereafter, after the period t5, as described above with reference to FIGS. 5B to 5E, the vertical transfer pulses are applied to the vertical transfer CCDs V11 to V28, and the horizontal transfer CCDs H11 to H22 are applied. By applying the horizontal transfer pulse, signal charges forming an even field image are taken out from the CCD image sensor 15.

  Such a readout method is referred to as a so-called “frame readout method”. First, signal charges accumulated in each of the light receiving portions R11 to R24 are alternately read out every other pixel. For example, in the operation described with reference to FIG. 5, first, the signal charges of the odd-numbered light receiving units are read out, and then the signal charges of the remaining even-numbered light receiving units are read out in the next field period. As a result, the reading interval in each light receiving unit is as long as twice as long as one field period (that is, one frame period), and thus a frame afterimage occurs. For this reason, in the frame readout method, the charge accumulation time (exposure time) is controlled by an electronic shutter function that ejects unnecessary charges by a reset pulse.

  In the operation example shown in the time charts of FIGS. 4A to 4D, the reset pulse is applied at t2 to t3 and t5 to t6 after the signal charge is shifted from each light receiving unit to the vertical transfer CCD. Thus, the signal charges accumulated in the respective light receiving portions so far are discarded, and the charge accumulation time of the odd field and the even field is limited to one field period. As a result, as shown in FIGS. 4A to 4D, periods t0 to t1 for accumulating signal charges constituting odd fields, and periods t3 to t4 for accumulating signal charges constituting even fields, , There is no time overlap.

  Therefore, the illumination pulse generation circuit shown in FIG. 2 generates a pulse signal that turns on the first pulse laser generator 21 that provides bright-field illumination during a period in which signal charges that form odd fields are accumulated, and a signal that forms even fields. During the period for accumulating charges, a pulse signal for turning on the second pulse laser generator 31 for applying dark field illumination is generated, so that an odd field that is a bright field image by bright field illumination and a dark field by dark field illumination. Both even fields that are images can be extracted from the CCD image sensor 15 in one frame.

In the operation example shown in the time charts of FIGS. 4A to 4F, periods t0 to t1 for accumulating signal charges forming odd fields and periods t3 to t3 for accumulating signal charges forming even fields are included. In a section not including t4 (section in which the signal is “1” in the chart of FIG. 4E), the first pulse laser generator 21 is turned on to provide bright field illumination.
At this time, for example, the illumination pulse generation circuit 25 detects the falling edge of the shift charge pulse in the even field and turns on the first pulse laser generator 21, and detects the rising edge of the shift charge pulse in the odd field and detects the first pulse. A pulse signal for turning off the laser generator 21 may be generated.

On the other hand, a period including signal periods t3 to t4 for accumulating signal charges constituting even fields and a period not including periods t0 to t1 for accumulating signal charges constituting odd fields (in the chart of FIG. 2), the second pulse laser generator 31 is turned on to provide dark field illumination.
At this time, for example, the illumination pulse generation circuit 25 detects the falling edge of the shift charge pulse in the odd field, turns on the second pulse laser generator 31, and detects the rising edge of the shift charge pulse in the even field. A pulse signal for turning off the laser generator 31 may be generated.

The odd field signal that is a bright field image read from the CCD image sensor 15 is converted into a digital image signal by the analog-digital conversion circuit 41 and input to the image processing unit 43, and the even field signal that is a dark field image is also converted. Then, it is converted into a digital image signal by the analog-digital conversion circuit 42 and input to the image processing unit 43.
The image processing unit 43 may output these odd field signals and even field signals as separate images and display them on a display device (not shown) or store them in a storage device (not shown). The images may be combined and output as a single image, displayed on a display device (not shown), or stored in a storage device (not shown).

In the above description, an example in which signal charges are read from the CCD image sensor 15 based on the frame reading method has been described. However, odd-numbered frames and even-numbered frames may be alternately read by the field reading method.
FIG. 6A is a diagram showing how an odd field is read from the CCD image sensor 15 by the field readout method, and FIG. 6B is a diagram showing how an even field is read by the field readout method.

In the field readout method, the accumulated charges of all the light receiving parts are read out in one field period, and the accumulated charges of the light receiving parts of two adjacent pixels are added and read. At this time, as shown in FIGS. 6A and 6B, the combination of columns to be added is changed between the odd field and the even field.
In the field readout method, since the signal accumulation time of each pixel is one field period, the signal charge constituting the odd field is accumulated without applying the reset pulse as shown in FIG. There is no time overlap between the periods for accumulating signal charges in the even field.

The first pulse laser generator 21 and the second pulse laser generator 31 may have different wavelengths of emitted laser light.
For example, when an inspection image used for visual inspection of a semiconductor wafer is obtained using the sample imaging apparatus 1, the first pulse laser generator 21 that provides bright field illumination resolves the bright field image. It is preferable to use illumination light with a short wavelength in order to improve the performance. However, in bright field illumination with such a short wavelength, the remaining titanium nitride or the like remaining between the aluminum wirings formed on the wafer surface is the surrounding aluminum. It is difficult to observe because it is hidden behind reflected light.

Therefore, in the dark field illumination, by using illumination light having a relatively longer wavelength than that of the bright field illumination so that the remainder such as titanium nitride between the aluminum wirings shines brightly, Detection can be facilitated.
At this time, the bright-field illumination light and the dark-field illumination light have different wavelengths, so that the focal position of the imaging optical system of the imaging apparatus 1 obtains the optimum position and dark-field image for obtaining a bright-field image. However, since the dark field image requires less resolution than the bright field image, the focal position of the imaging optical system of the imaging apparatus 1 is the same as that of the bright field illumination light. It is preferable to adjust according to the wavelength.

FIG. 7 is an overall configuration diagram showing an outline of the sample imaging device 1 according to the second embodiment of the present invention. In this configuration, the laser generator 26 that provides bright-field illumination and dark-field illumination is shared, and an optical switch 51 such as an acousto-optic switch is used for either the bright-field illumination optical system or the dark-field illumination optical system. The illumination of the sample imaging device 1 is switched between bright-field illumination and dark-field illumination by switching whether to guide the illumination light generated by the laser generator 26.
In this configuration example, the illumination pulse generation circuit 25 detects the falling edge of the shift pulse of the signal charge forming the even field, switches the output destination of the laser generator 26 to the bright field illumination optical system, and sets the odd field. A falling edge of the signal charge shift pulse is detected, and a switching signal for switching the output destination of the laser generator 26 to the dark field illumination optical system is generated and output to the optical switch 25.

  In the sample imaging device 1 shown in FIGS. 2 and 7, the stage 11 is moved to move the CCD image sensor 15 relative to the sample 2, and on the light receiving surface of the CCD image sensor 15 accompanying this relative movement. Each signal charge taken out to the CCD from each light receiving unit arranged in the moving direction at a speed corresponding to the moving speed of the formed image is transferred in the moving direction of the image and arranged in the moving direction. Alternatively, the CCD image sensor 15 may be used as a TDI (time delay integration) line sensor by superimposing signal charges taken out from the respective light receiving portions on the CCD.

FIGS. 8A to 8F show a CCD for superimposing the odd-field signal charges and the even-field signal charges separately when the CCD image sensor 15 is used as a TDI line sensor. FIG. 9 is a time chart showing a reading operation of the image sensor 15, and FIGS. 9A to 9F are explanatory diagrams of such a reading operation.
FIG. 8A shows the timing of applying the shift charge pulse to the vertical transfer CCDs V12, V16, V22, and V26 that take out signal charges from the odd-numbered light receiving units R11, R13, R21, and R23, respectively. The signal charges accumulated in the odd-numbered light receiving portions R11, R13, R21, and R23 are shifted during the period “1”.

  FIG. 8B shows the timing of applying the shift charge pulse to the vertical transfer CCDs V14, V18, V24 and V28 for extracting the signal charges from the even-numbered light receiving units R12, R14, R22 and R24, respectively. In the period, the signal charges accumulated in the even-numbered light receiving portions R12, R14, R22, and R24 are shifted.

FIG. 8C shows a period in which vertical transfer pulses and horizontal transfer pulses are applied to the vertical transfer CCDs V11 to V28 and horizontal transfer CCDs H11 to H22, respectively. In the period “1” shown in FIG. The accumulated signal charges are transferred in the vertical direction by two pixels. As a result, the signal charges transferred to the horizontal transfer CCD are transferred in the horizontal direction and taken out as output signals.
FIG. 8D shows the timing at which the reset pulse is applied. The signal charges accumulated in the light receiving portions R11 to R24 during the period “1” shown in the figure are ejected toward the substrate and erased.

  First, at time t2, when a shift charge pulse is applied to the vertical transfer CCDs V12, V16, V22, and V26 by applying a high charge φV2, the signal charges accumulated in the odd-numbered column light receiving units R11, R13, R21, and R23 are vertical. The transfer CCDs are shifted to V12 and V11, V16 and V15, V22 and V21, and V26 and V25, respectively (see FIG. 9A).

  At time t4, when a high potential φV4 is applied to the vertical transfer CCDs V14, V18, V24 and V28 and a shift charge pulse is applied, the signal charges accumulated in the even-numbered light receiving units R12, R14, R22 and R24 are vertically transferred. Shifted to CCDV14 and V13, V18 and V17, V24 and V23, and V28 and V27, respectively (see FIG. 9B).

  Thereafter, at time t5 to t6, a vertical transfer pulse is applied to each of the vertical transfer CCDs V11 to V28 to transfer the signal charge accumulated in each vertical transfer CCD by two pixels in the vertical direction ((C) in FIG. 9). The signal charges transferred to the horizontal transfer CCD during this period are taken out as output signals by applying horizontal transfer pulses to the horizontal transfer CCDs H11 to H22.

At time t7 again, a shift charge pulse is applied to the vertical transfer CCDs V12, V16, V22, and V26, so that the signal charges accumulated in the odd-numbered column light receiving units R11, R13, R21, and R23 are vertically transferred CCDV12, V16, V22. And V26 are respectively superposed on the signal charges previously transferred to these vertical transfer CCDs (see FIG. 9E).
Further, by applying a shift charge pulse to the vertical transfer CCDs V14, V18, V24 and V28 at time t8, the signal charges accumulated in the even-numbered column light receiving units R12, R14, R22 and R24 are transferred to the vertical transfer CCDs V14, V13 and V18. Are taken out to V17, V24 and V23 and superposed on the signal charges previously transferred to these vertical transfer CCDs (see FIG. 9F).

By repeating the above operations, the signal charges photoelectrically converted by the odd-numbered light receiving units R11, R13, R21, and R23 and the signal charges photoelectrically converted by the even-numbered light receiving units R12, R14, R22, and R24 are individually provided. Superimposed.
Here, the stage control unit 13 moves the image of the sample 2 formed on the light receiving surface in the same direction as the signal charge moving direction by the vertical transfer CCD, and the odd-numbered light receiving units R11, R13, R21, and R23. The stage 11 is moved so that the image of the sample 2 moves by two pixels during an interval ΔT (time t2 to t7 or time t4 to t8) at which the signal charges accumulated in the vertical transfer CCD are shifted.
Thus, by transferring the signal charge accumulated in the vertical transfer CCD at a speed corresponding to the moving speed of the image formed on the light receiving surface, the image of the sample 2 imaged on the light receiving surface Photoelectric signals obtained by capturing light reflected from the same portion by a plurality of light receiving units are superimposed on the same vertical transfer CCD, and the CCD image sensor 15 functions as a TDI line sensor.

  Similar to the time chart of the frame reading operation described with reference to FIGS. 4A to 4D, as shown in FIG. A reset pulse is applied at time t1, time t3,. As a result, the charge accumulation time of the odd field shifted to the CCD at time t2 becomes time t1 to t2, and the charge accumulation time of the even field shifted to the CCD at time t4 becomes time t3 to t4.

  FIG. 8E is a time chart for providing bright field illumination from the above-described bright field illumination optical system, and shows that bright field illumination is applied in a section where the signal is “1”. ) Is a time chart for providing dark field illumination from the above-described dark field illumination optical system, and indicates that dark field illumination is applied in a section where the signal is “1”. As shown in FIGS. 8E and 8F, bright field illumination is given during the charge accumulation time (t1 to t2) in the odd field, and dark field illumination is given during the charge accumulation time (t3 to t4) in the even field. By switching to, a TDI image by bright field illumination and a TDI image by dark field illumination can be simultaneously imaged in parallel as an odd field and an even field, respectively.

For example, in the configuration example shown in FIG. 2, the illumination pulse generation circuit 25 causes the first pulse laser generator 21 to generate illumination light between times t1 and t2 as shown in FIG. By generating a pulse signal that causes the second pulse laser generator 31 to generate illumination light between t4 and t4, bright field illumination and dark field illumination are alternately switched and generated.
In this way, in order to switch the first pulse laser generator 21 and the second pulse laser generator 31 alternately and light up, the illumination pulse generation circuit 25 detects any pulse laser when the falling edge of the reset pulse is detected. When the generator is turned on and the rising edge of the shift pulse to the CCD image sensor 15 is detected, a pulse signal for turning off the previously turned on pulse laser generator may be generated. At this time, the illumination pulse generation circuit 25 may alternately generate a pulse signal for lighting the first pulse laser generator 21 and a pulse signal for lighting the second pulse laser generator 31.

  Further, for example, in the configuration example shown in FIG. 7, the illumination pulse generation circuit 25 causes the pulse laser generator 26 to emit light when detecting the falling edge of the reset pulse, and detects the rising edge of the shift pulse to the CCD image sensor 15. Sometimes, a pulse signal for turning off the pulse laser generator 26 may be given to the pulse laser generator 26. At this time, the illumination pulse generation circuit 25 generates a switching signal for alternately switching the output destination of the laser generator 26 between the bright field illumination optical system and the dark field illumination optical system, and outputs it to the optical switch 25. It is good.

  INDUSTRIAL APPLICABILITY The present invention can be used for an imaging apparatus that images a sample illuminated by bright field illumination and dark field illumination, and a sample illumination method. In particular, the present invention can be suitably used for an imaging apparatus and a sample illumination method used in an appearance inspection apparatus that inspects a pattern formed on a sample surface such as a semiconductor wafer in a semiconductor manufacturing process or the like.

(A) is a figure which shows the schematic structural example of a bright field illumination optical system, (B) is a figure which shows the schematic structural example of a bright field illumination optical system. 1 is an overall configuration diagram showing an outline of a sample imaging apparatus according to a first embodiment of the present invention. It is a block diagram of a part of CCD image sensor. 4 is a time chart showing an interlace reading operation in the CCD image sensor shown in FIG. 3. It is explanatory drawing of the read-out operation | movement at the time of reading the signal charge of the CCD image sensor shown in FIG. 3 by a frame read-out system. It is explanatory drawing of the read-out operation | movement at the time of reading the signal charge of the CCD image sensor shown in FIG. 3 by a field read-out system. It is a whole block diagram which shows the outline of the sample imaging device by 2nd Example of this invention. It is a time chart which shows the read-out operation | movement at the time of using the CCD image sensor shown in FIG. 3 as a TDI sensor. It is explanatory drawing of the read-out operation | movement at the time of using the CCD image sensor shown in FIG. 3 as a TDI sensor.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Sample imaging device 2 Sample 11 Stage 12 Sample stand 14 Objective lens 15 CCD image sensor 21, 31 Pulse laser generator 22, 23, 32, 33 Illumination lens 24 Semi-transparent mirror 34 Perforated mirror 35 Mirror

Claims (10)

An image sensor in which pixel signals forming odd fields and pixel signals forming even fields are alternately extracted from the light receiving unit;
An illumination light switching unit that switches illumination light that illuminates the sample surface in synchronization with switching of a field from which the pixel signal is extracted from the light receiving unit;
A sample imaging device comprising:   The sample imaging apparatus according to claim 1, wherein the illumination light switching unit switches between bright field illumination and dark field illumination.   The sample imaging apparatus according to claim 2, wherein the bright field image by the bright field illumination and the dark field image by the dark field illumination are both formed on a light receiving surface of the imaging element.   4. The sample imaging device according to claim 2, wherein the bright field illumination and the dark field illumination have different wavelengths of illumination light.   The sample imaging apparatus according to claim 1, wherein the imaging element is a CCD image sensor. A moving mechanism for moving the image sensor relative to the sample;
The CCD image sensor is taken out to the CCD from each light receiving unit arranged in the moving direction at a speed corresponding to the moving speed of the image formed on the light receiving surface of the image pickup device with the relative movement. Transferring each pixel signal in the moving direction of the image,
The sample imaging device according to claim 5. A method for illuminating a sample in an imaging device that images a sample surface using an imaging device in which pixel signals forming an odd field and pixel signals forming an even field are alternately extracted from a light receiving unit,
An illumination method, wherein illumination light for illuminating a sample surface is switched in synchronization with switching of a field for extracting the pixel signal from the light receiving unit.   The illumination method according to claim 7, wherein bright field illumination and dark field illumination are switched in synchronization with the field switching.   The illumination method according to claim 8, wherein a bright field image by the bright field illumination and a dark field image by the dark field illumination are both formed on the imaging device.   The illumination method according to claim 8 or 9, wherein a wavelength of illumination light is different between the bright field illumination and the dark field illumination.

Images