APPARATUS AND METHOD FOR INSPECTING DEFECTS
A defect inspection apparatus includes a movable stage for mounting a substrate having circuit patterns as an object of inspection, an irradiation optical system which irradiates a slit-shaped light beam from an oblique direction to the circuit patterns of the substrate, a detection optical system which includes an image sensor for receiving reflected/scattered light from the substrate by irradiation of the slit-shaped light beam and converting the received light into a signal, and an image processor which processes the signal. The irradiation optical system includes a cylindrical lens and a coherency reduction optical system, which receives the light beam and emits a plurality of slit-shaped light sub-beams which are spatially reduced in coherency in a light-converging direction of the cylindrical lens. The cylindrical lens focuses the plurality of slit-shaped light sub-beams into the slit-shaped light beam irradiated to the surface of the substrate.
This application is a continuation application of U.S. application Ser. No. 12/328,357, filed Dec. 4, 2008, now allowed, which is a continuation of U.S. application Ser. No. 12/029,660, filed Feb. 12, 2008, now abandoned, which is a continuation application of U.S. application Ser. No. 11/653,322, filed Jan. 16, 2007, now U.S. Pat. No. 7,333,192, which claims the benefit of Japanese Application No. JP 2006-013285, filed Jan. 23, 2006 in the Japanese Patent Office, the disclosures of which are incorporated herein by reference.
BACKGROUND OF THE INVENTIONThe present invention relates to an apparatus and method for inspecting defects, adapted to detect and analyze contamination and other defects occurring during the manufacturing processes of a desired object that include forming patterns on a substrate, to provide appropriate measures against contamination and other defects, and to inspect the occurrence status thereof. The manufacturing processes include semiconductor manufacturing processes, liquid-crystal display device manufacturing processes, printed-circuit board manufacturing processes, and the like.
During conventional semiconductor-manufacturing processes, the presence of dust particles or other foreign substances (hereinafter, referred to collectively as contamination) on the surface of a semiconductor substrate (wafer) causes defects such as the improper insulation or short-circuiting of wiring. In addition, if a semiconductor device becomes fine-structured to form fine-structured contamination in the semiconductor substrate, the formation of the contamination results in the improper insulation of capacitors and/or in damage to gate oxide films or the like. Such contamination gets entrapped in various forms for various reasons. For example, the contamination may have stemmed from the movable section of a transport device, from the human body, or from an internal reaction product of a processing device due to a process gas, or may have been included in a chemical or a material. The occurrence of contamination or some other defect on patterns during liquid-crystal display device manufacturing processes similar to the above manufacturing processes renders the display device inoperative. The same situation also arises during printed-circuit board manufacturing processes; the presence of contamination causes pattern short-circuiting and/or improper connection.
A known technique for inspecting the above-mentioned contamination is by irradiating a wafer with coherent light, spatially filtering away the light exited from an iterative pattern present on the wafer, and detecting non-repeatable contamination or other defects in edge-enhanced form. Also, Japanese Laid-Open Patent Application Publication (JP Kokai) No. Hei 01-117024 (Patent Document 1) discloses a contamination inspection apparatus constructed such that a circuit pattern formed on a wafer is irradiated from a direction inclined at 45 degrees to a major straight-line group of the circuit pattern to prevent the zeroth-order diffracted light from the major straight-line group from being input to an aperture of an objective lens. It is also described that in the conventional technique of Patent Document 1, straight lines not belonging to the main straight-line group are light-shielded with a spatial filter. In addition, other known apparatuses and methods for inspecting contamination and other defects are described in JP Kokai Nos. Hei 06-258239 (Patent Document 2), Hei 08-210989 (Patent Document 3), 2000-105203 (Patent Document 4), 2004-93252 (Patent Document 5), 2004-177284 (Patent Document 6), and 2004-184142 (Patent Document 7). That is to say, Patent Documents 2 and 7 describe a technique using a pitch-variable spatial filter. Patent Document 3 describes a technique in which, after a plurality of light fluxes spatially incoherent with each other have been obtained by splitting a beam emitted from a laser source and combining mirrors for the fluxes to differ from one another in optical path length, a desired section on a substrate is irradiated with the obtained plurality of fluxes obliquely at mutually different angles of incidence and then the reflected/scattered light arising from very small defects on the substrate during the irradiation is converged via detection optics and received with photoelectric conversion means. Patent Document 7 further describes a technique using various forms of spatial filters, inclusive of a transmissive type of liquid-crystal filter. Patent Documents 4, 5, and 6 describe an apparatus for inspecting defects such as contamination, equipped with highly efficient illumination optics to emit a slit-shaped beam of illumination light from a direction of 45 degrees for reduced incidence of pattern-scattered light on an objective lens.
SUMMARY OF THE INVENTIONIn Patent Documents 1-7 outlined above, however, sufficient consideration has not been paid to a design capable of inspecting very small particle-like contamination, thin-film-like contamination, scratches, and other defects present on various types of substrates that is to be inspected, in a simple configuration. The various types of substrates includes substrates having various regions (such as a region with the rough edges of an ultrafine-structured circuit pattern, a region with an iterative pattern and a non-iterative pattern in mixed form, a region different in pattern density from other regions, and a region with a transparent film).
In order to solve the above problem, the present invention provides an apparatus and method for inspecting defects, constructed such that very small particle-like contamination, thin-film-like contamination, scratches, and other defects present on various types of substrates to be inspected can be inspected in a simple apparatus configuration at high speed and with high accuracy. The various types of substrates include substrates with various regions.
The present invention also provides an apparatus and method for inspecting defects, adapted such that a highly efficient manufacturing line for substrates can be constructed by using the apparatus as an in-line monitor.
An aspect of the present invention is an apparatus and a method for inspecting defects, the apparatus including: a stage that moves with a substrate (to be inspected) being rested thereon, the substrate including a circuit pattern formed thereon; an irradiation optical system that guides a flux of beams emitted from a laser light source, from a principal optical path to a first or second optical path, then focuses the beam flux that has been guided to the first or second optical path into a slit-shaped beam formed of fluxes of light that are substantially parallel in a longitudinal direction of the slit-shaped beam, from a direction having a required inclination extending horizontally with respect to a major straight-line group of the circuit pattern and at a required angle of inclination to the surface of the substrate, and irradiates the substrate with the slit-shaped beam such that the longitudinal direction is at a substantially right angle to a traveling direction of the stage; a detection optical system that uses an objective lens to converge reflected/scattered light obtained from contamination or defects present on the substrate that has been irradiated with the slit-shaped beam, and uses an image sensor to receive the reflected/scattered light that has been converged, convert the received light into a signal, and detect the signal; and an image processor which, on the basis of the signal that has been detected by the image sensor of the detection optical system, extracts a signal indicative of the defect such as contamination.
The irradiation optical system in the above apparatus and method for inspecting defects includes a mirror that reflects and directs downward the beam flux that was guided to the first or second optical path, and a cylindrical lens and inclined mirror for focusing the beam flux that has been directed downward by the foregoing mirror onto the substrate, from the direction having the required inclination extending horizontally and at the required inclination angle, and thus obtaining the slit-shaped beam.
Also, the inclined mirror in the above irradiation optical system is constructed to make the slit-shaped beam selectively usable for irradiation at both a high angle of inclination and a low angle of inclination.
Another aspect of the present invention is an apparatus and method for inspecting defects, the apparatus including: a stage which is caused to travel after a substrate to be inspected has been rested on the stage, the substrate including a circuit pattern formed thereon; an irradiation optical system which, after emitting a flux of beams from a laser light source, guides the beam flux from a principal optical path to a first optical path, a second optical path, and a third optical path each, then focuses the beam flux that has been guided as a first beam flux to the first optical path into a first slit-shaped beam, from a clockwise direction having a required inclination extending horizontally with respect to an X-axis direction of a major straight-line group of the circuit pattern and at a required angle of inclination to the surface of the substrate to be inspected, the first slit-shaped beam being formed of fluxes substantially parallel in a longitudinal direction, and irradiates the substrate with the first slit-shaped beam such that the longitudinal direction is substantially at right angles to a traveling direction of the stage, focuses the beam flux that has been guided as a second beam flux to the second optical path into a second slit-shaped beam from a counterclockwise direction having a required inclination extending horizontally with respect to the X-axis direction and at a required angle of inclination to the surface of the substrate to be inspected, the second slit-shaped beam being formed up by fluxes substantially parallel in a longitudinal direction, and irradiates the substrate with the second slit-shaped beam such that the longitudinal direction is at a substantially right angle to the traveling direction of the stage, and focuses the beam flux that has been guided as a third beam flux to the third optical path into a third slit-shaped beam, from the X-axis direction extending horizontally and at a required angle of inclination to the surface of the substrate, the third slit-shaped beam being formed of fluxes substantially parallel in a longitudinal direction, and irradiates the substrate with the third slit-shaped beam such that the longitudinal direction is at a substantially right angle to the traveling direction of the stage; a detection optical system which uses an objective lens to converge reflected/scattered light obtained from contamination or some other defect present on the substrate that has been irradiated with the slit-shaped beam, the detection optical system further using an image sensor to receive the reflected/scattered light that has been converged, convert the received light into a signal, and detect the signal; and an image processor which, on the basis of the signal that has been detected by the image sensor of the detection optical system, extracts a signal indicative of the defect such as contamination.
The irradiation optical system in the above apparatus and method for inspecting defects includes: a first mirror which reflects and directs downward the first beam flux that has been guided to the first optical path; a first cylindrical lens and a first inclined mirror for focusing the beam flux that has been directed downward by the first mirror onto the substrate, from a clockwise direction having the required inclination extending horizontally and at the required inclination angle, and thus obtaining the first slit-shaped beam; a second mirror which reflects and directs downward the second beam flux that has been guided to the second optical path; a second cylindrical lens and a second inclined mirror for focusing the beam flux that has been directed downward by the second mirror onto the substrate, from the counterclockwise direction having the required inclination extending horizontally and at the required inclination angle, and thus obtaining the second slit-shaped beam; a third mirror which reflects and directs downward the third beam flux that was guided to the third optical path; and a third cylindrical lens and a third inclined mirror for focusing the beam flux that has been directed downward by the third mirror onto the substrate, from the X-axis direction extending horizontally and at the required inclination angle, and thus obtaining the third slit-shaped beam.
Also, the irradiation optical system in the above apparatus and method for inspecting defects is constructed such that principal-beam expanding optics for expanding beams in diameter is disposed in the principal optical path and such that beam-expanding optics for expanding the third beam flux further in diameter and matching the first, second, and third slit-shaped beams in longitudinal length on the substrate is disposed in the third optical path.
In addition, the irradiation optical system in the above apparatus and method for inspecting defects is constructed such that an optical member group formed by stacking a plurality of plate-shaped optical members each having a different optical path length in at least a light-converging direction of the second cylindrical lens is disposed on the first, second, or third optical path to receive a coherent second beam flux obtained from the principal optical path, and emit a plurality of slit-shaped beams each spatially reduced in coherency, in the light-converging direction of at least the second cylindrical lens.
Furthermore, the irradiation optical system in the above apparatus and method for inspecting defects is constructed such that beam-expanding optics for expanding the incident beam flux in diameter in at least a stacking direction of the plurality of plate-shaped optical members is disposed at an incident side of the optical member group.
Further, the irradiation optical system in the above apparatus and method for inspecting defects is constructed such that an optical member group formed by stacking a plurality of plate-shaped optical members each having a different optical path length in at least a light-converging direction of a cylindrical lens provided on at least a selected path is disposed on the principal optical path so as to be movable in and out with respect thereto for receiving a coherent beam flux obtained from the laser light source, and emit a plurality of slit-shaped beams each spatially reduced in coherency, in the light-converging direction of the cylindrical lens.
Furthermore, the first, second, and third inclined mirrors in the above irradiation optical system are constructed so as to allow the slit-shaped beams to be switched for irradiation at a high angle of inclination and irradiation at a low angle of inclination.
Furthermore, the detection optical system in the above apparatus and method for inspecting defects is constructed such that a spatial filter for light-shielding an interference pattern formed from an iterative circuit pattern present on the substrate is disposed at a position conjugate to a pupil of the objective lens.
Further, the irradiation optical system in the above apparatus and method for inspecting defects is constructed such that a luminous quantity adjusting filter for adjusting the amount of light is disposed on the principal optical path. The irradiation optical system in the above apparatus and method for inspecting defects is further constructed such that a polarizing plate for conducting polarization control is disposed on the principal optical path, and the detection optical system is constructed to include an analyzer that controls and detects polarization.
Moreover, the laser light source in the above apparatus and method for inspecting defects is constructed to emit UV or DUV laser light. The above apparatus and method for inspecting defects is further constructed to make a magnification of detected image formation variable in the detection optical system.
These and other objects, features and advantages of the invention will be apparent from the following more particular description of preferred embodiments of the invention, as illustrated in the accompanying drawings.
Hereunder, embodiments of the present invention will be described using the accompanying drawings.
One possible kind of object 1 to be subjected to inspection of defects such as contamination, is a semiconductor wafer “1a” with chips “1aa” arrayed two-dimensionally at required spatial intervals (hereinafter, the object 1 is referred to as the inspection target 1). An interior of each chip “1aa” is mainly formed up by a register group region “1ab”, a memory region “1ac”, a CPU core region “1ad”, and an input/output region “1ae”. The register group region “1ab” and the memory region “1ac” are both formed by two-dimensionally and regularly arranging patterns with a minimum line width ranging from about 0.05 to 0.02 μm. The CPU core region “1ad” and the input/output region “1ae” are both formed by non-iteration of patterns with a minimum line width ranging from about 0.05 to 0.02 μm. In this way, even when the inspection target 1 for the inspection of contamination and other defects is a semiconductor wafer and chips are regularly arrayed thereon, the inspection target 1 may take various forms; the interior of each chip varies from region to region in terms of minimum line width, patterns are created in iterative or non-iterative form, and/or transparent thin films are formed.
Embodiments of the present invention relating to an apparatus and method for inspecting defects such as contamination are adapted to allow highly-sensitive and high-speed inspection of very small defects ranging from 0.05 μm to 0.10 μm in size, such as contamination, scratches, and pattern defects, that are present on the inspection target 1 which takes various forms as mentioned above. In the above defect inspection apparatus and method, defects can also be classified according to the type thereof.
A first embodiment of an apparatus for inspecting contamination and other defects according to the present invention is described below using
The first embodiment of an apparatus for inspecting defects such as contamination includes a transport system (stage system) 10, an irradiation optical system 20, a detection optical system 30, an image processor 40, an observation optical system 50, an in-focus control system 60, a total controller 70, and a reviewing optical microscope 80. The transport system (stage system) 10 includes an X-stage 11, Y-stage 12, and Z-stage 13, each for resting thereon and moving the inspection target substrates 1, such as wafers, that are obtained from various product types and various manufacturing process steps. The transport system (stage system) 10 also includes a θ-stage 14 (rotary stage in a horizontal plane) and a stage controller 15.
The irradiation optical system 20 includes a laser light source 21, a luminous quantity adjusting filter (luminous attenuation filter) 22 for adjusting the amount of light, a polarizing plate 23 for converting the light into, for example, linearly polarized P- or S-components, principal-beam expanding optics 24 for expanding a beam flux to a certain diameter, and light-converging optics 25 that uses a mirror group 26 and/or other optical members to converge the beam flux upon the substrate 1 in the form of a slit-shaped beam elongated in a direction perpendicular to a traveling direction of the substrate. The irradiation optical system 20 further includes an illumination controller 27 that controls or selects an illumination angle and a filter type under the commands sent from the total controller 70. In consideration of highly sensitive inspection of contamination and other defects and of minimum maintenance costs, a second-harmonic generation (SHG) type of high-output YAG laser with a 532-nm wavelength is preferably used as the laser light source 1. However, the laser light source does not always need to provide 532 nm in wavelength and may emit an ultraviolet (UV) laser, a far-ultraviolet (FUV) or deep-ultraviolet (DUV) laser, a vacuum UV laser, an argon (Ar) laser, a nitrogen laser, a helium-cadmium (He—Cd) laser, an excimer laser, a semiconductor laser, or the like. An advantage obtained from using any of these laser light sources is that since shortening the laser wavelength enhances resolution of the images detected, defects can be inspected with higher sensitivity. For a short wavelength of about 0.34 μm, an objective lens 31 preferably has a numerical aperture (NA) of about 0.4, or for a wavelength of about 0.17 μm, an NA of about 0.2. Thus, a large amount of diffracted light from a defect such as contamination can be made to enter the objective lens 31 for improved detection sensitivity.
The detection optical system 30 includes, as shown in
The image processor 40 includes: an ND converter 41 for ND conversion of an image signal detected by the photodetector 36; a data storage unit 42 for storing the detected image signal “f(i, j)” that has been obtained by the ND conversion; a threshold calculation processor 43 for calculating threshold levels based on the detected image signal; a contamination detection processor 44 for conducting contamination detection processing for each pixel-merging operation, based on the detected image signal obtained from the data storage unit 42 and on the threshold image signal obtained from the threshold calculation processor 43; a feature calculating circuit 45 for calculating features for each merging operation, such as the amount of scattered light that has been obtained from defect detection, and the quantity (area, projection length, or like value) of detected pixels, indicating a spread of the defect; and an integrated processor 46 that classifies on-wafer small/large contamination, pattern defects, microscratches, and other defects into various types of defects on the basis of the features obtained from the feature calculating circuit 45 for each merging operation. Inspection results obtained by the integrated processor 46 are then supplied to the total controller 70 for display in a GUI screen of a display unit 72.
The observing optical system 50 includes a mirror 51 disposed on a detection optical path so as to be movable in and out with respect thereto, as indicated by arrows; an optical branching element 52 such as a half-mirror; observation optics 53 for observing a light-shielding pattern of a spatial filter 32 provided on a Fourier transform plane; and observation optics 54 for observing an image of the wafer surface. The observation optics 53 includes an image-forming lens 531 and an image sensor 532 such as a TV camera. The observation optics 54 also includes an imaging lens 541 and an image sensor 542 such as a TV camera.
The in-focus control system 60 includes: a light source 61 such as a lamp; a striped pattern (grid-shaped pattern) 62 to be projected onto the surface of the wafer by using parallel light fluxes that are emitted from the light source 61; a half-mirror 63; a projection lens 64 for projecting the striped pattern onto the wafer surface; an imaging lens 65 for forming an image of the striped pattern regularly reflected from the wafer surface; a mirror 66 for reflecting the striped pattern formed by the imaging lens 65; an imaging lens 67 that forms an striped pattern obtained through the projection lens 64 and the half-mirror 63 after the striped pattern formed by the imaging lens 65 is reflected by the mirror 66 and returned to parallel light fluxes by the imaging lens 65 to being regularly reflected on the wafer surface; a one-dimensional or two-dimensional linear image sensor 68 for detecting, as a positional shift of the striped pattern, an out-of-focus state on the wafer surface; and an in-focus controller 69 for controlling, for example, the Z-stage 13 according to the out-of-focus state detected by the linear image sensor, and conducting an in-focus matching process on the surface of the wafer.
The total controller 70 includes: an input device 71 (including a keyboard, a mouse, a network interface, and so on) for entering information on the various wafers loaded; the display unit 72 that has modes such as a GUI screen mode for displaying the image acquired by the image sensor 36, the screen image observed through the observation optics 53, 54, and images of the contamination and other defects reviewed through the reviewing optical microscope 80, and displays the threshold level information calculated by the threshold calculation processor 43 or the like, and the inspection results and other data obtained in the integrated processor 46; and a data storage unit 73 for storing various information such as CAD information of the wafers, inspection parameter information on past wafers, and the inspection result information obtained in the integrated processor 46.
The reviewing optical microscope 80 includes a light source 81 for emitting visible light such as white light, a half-mirror 82, an objective lens 83, an imaging lens 84, and an image sensor 85. The microscope 80 acquires images of defects or false defects or the like and supplies the images to the total controller 70, from which the images are then transferred to the display unit 72 for display and classification of each image in GUI screen mode. Thus, information on position coordinates of the defects is acquired and review results are acquired. That is to say, when the stages 11, 12 are moved, the reviewing optical microscope 80 observes images of the contamination or defects (including false defects) on a wafer 1, detected by the image processor 40 of the defect inspection apparatus and stored into the storage unit 73, for example. Next, the microscope 80 acquires review images of the various forms of contamination or defects shown as 3903 in
Next, specific examples of the irradiation optical system 20 which is a feature of the present invention are described below using
The first inclined mirror 2603a is constructed such that its inclination α1 can be changed to a low angle of about 5-15 degrees and such that its inclination α2 can be changed to a high angle of about 40-55 degrees.
Since, as described above, the slit-shaped beam 90 is irradiated from the direction having the inclination angle φ with respect to the X-axis direction, the zeroth-order diffracted light arising from the major straight-line group can be prevented from entering the aperture of the objective lens 31. In addition, if the patterns formed on the wafer are contact holes, capacitors, or the like, since these elements have no specific directivity, the beam is preferably irradiated from an oblique direction with the angle φ of about 45 degrees.
The coherency reduction optical system 255 including the beam-expanding optics 2551 and the optical member group 2552 is built into the counterclockwise θ-angled irradiation optical system 200b, that is, the second optical path 221b. The coherency reduction optical system 255 may be provided in the first optical path 221a or the third optical path 221c if installation space requirements permit this configuration. However, it is possible to install the coherency reduction optical system 255 in any other optical path if this path is formed so as to be insertable into and retreatable from the first optical path 221a or the third optical path 221c or selectable as a shunt to replace either of the latter two paths.
Alternatively, if the principal optical path 220 is constructed to be installable, the coherency reduction optical system 255 may be provided between the principal-beam expanding optics 24 forming the principal optical path 220, and the optical elements 2610-2612. In this case, any one of the first optical path 221a, the second optical path 221b and the third optical path 221c can be used to irradiate a beam of reduced coherency, as the slit-shaped beam 90.
As described above, slit-shaped beam 90, 90′ can be irradiated in any one of five ways. In addition, in consideration of using two different values as the inclination angles α and β each, 10 ways are available to irradiate the beam.
The present invention is characterized in that, in the clockwise φ-angled irradiation optical system 200a and the counterclockwise φ-angled irradiation optical system 200b, the same functions as those of the conical lenses described in JP Kokai Nos. 2000-105203 and 2004-177284 are implemented using the cylindrical lens 251 and the switchable inclined mirror 2604. As can be understood from this, since special optical components are not used, the above optical elements (optical components) 251, 2602-2604 can be manufactured easily, and since the cylindrical lens 251, in particular, has a predetermined focal length, adjustment of these elements is easy and extensive cost reduction is possible.
In addition, in the irradiation optical system 20 according to the present invention, if, as shown in
Furthermore, in the irradiation optical system 20 according to the present invention, the laser beam emitted from the laser light source 21 has a Gaussian distribution as shown in
Selection of an inclination angle (elevation) α, β by means of the switchable inclined mirror 2604, 2624, is described below using
Next, another example of a detection optical system is described below using
As an alternative measure, the detection optical systems 30 and 30′ shown in
When an iterative pattern is irradiated with the slit-shaped beam 90, 90′, such diffracted light from the iterative pattern as shown in
Two methods are available to clear simultaneously the diffracted light patterns from the plural kinds of pattern regions by conversion into a Fourier transform image. One possible method is by inserting an irradiation range-limiting diaphragm into an optical path of the irradiation optical system to prevent the slit-shaped beam 90, 90′ from being irradiated spanning different kinds of pattern regions, or by making the beam-expanding optics 24 movable and modifying a beam-expanding magnification for a limited irradiation range. The other possible method is by providing a diaphragm in order for the diffracted light incident on a Fourier transform plane to be restricted into diffracted light of a smaller angle of incidence to reduce an N.A. value of the Fourier transform lens 31 used.
Selecting either of the above methods makes it possible for only the diffracted light pattern from a specific pattern region to be formed on the Fourier transform plane. Consequently, a light-shielding pattern appropriate for the diffracted light pattern from the specific pattern region can be set as the spatial filter 32, and the defects in the specific pattern region can be inspected with high sensitivity.
The spatial filter 32 may otherwise be constructed using a pitch-variable filter 160 constituted by plate-shaped or rod-shaped optical shielding members 163 arrayed between springs 161 and 162 coiled in mutually inverse directions, as shown in
Further alternatively, the spatial filter 32 may be constructed using a liquid-crystal filter 180 capable of being formed with any pattern, as shown in
The spatial filter 32 can otherwise be a filter having multiple kinds of light-shielding patterns printed thereon so as to be switchable from one pattern to another.
Next, a description is given below of an example of irradiating obliquely a polarized slit-shaped beam using the polarizing plate 23, and an example of detecting polarized light using the analyzer (polarizing filter) in the optical filter group 34.
During S-polarized irradiation of a slit-shaped beam 90 using the polarizing plate 23, S-polarized components are returned from the iterative pattern, whereas a mixture of S-polarized components and P-polarized components is returned from the defect such as contamination. Using the analyzer (polarizing filter) 34 to light-shield the S-polarized components, therefore, makes it possible for the image sensor 36 to receive the P-polarized components obtained from the defect such as contamination. Thus, a signal identifying the defect can be detected and the top layer defect can be detected with high sensitivity.
During P-polarized irradiation of the slit-shaped beam 90 via the polarizing plate 23, S-polarized components are returned from the iterative pattern, whereas a mixture of S-polarized components and P-polarized components is returned from the defect such as contamination. Using the analyzer (polarizing filter) 34 to light-shield the P-polarized components, therefore, makes it possible for the image sensor 36 to receive the S-polarized components obtained from the defect such as contamination. Thus, a signal identifying the defect can be detected and the underlayer defect can be detected with high sensitivity.
Next, the transport system 10 is described below. The stages 11, 12 move a sample placement table 14 along an XY plane and have a function that moves the entire water (inspection target substrate) 1 to an irradiation area of the irradiation optical system 20. The stage 13 is a Z-stage having a function which moves the sample placement table 14 in the optical-axis direction (Z-direction) of the detection optical system 30. The sample placement table 14 holds the wafer 1 and has a function that rotates the wafer 1 in a horizontal direction. The stage controller 15 has a function that controls the stages 11, 12, 13, and the sample placement table 14.
Next, a more specific example of an in-focus control system 60 is described below using
Autofocusing can therefore be realized by, for example, controlling the Z-stage 13 in accordance with the signal obtained from the linear sensor 68 according to the out-of-focus state.
Next, an example of a process sequence of inspecting defects such as contamination is described below using
The above is followed by step S295, in which, while the stages 11-14 are being moved, a position of an alignment mark formed on the wafer is detected using wafer alignment optics (not shown) and the detected position is matched to a reference position to perform the wafer alignment.
Next, the process proceeds to the inspection of defects such as contamination. During the defect inspection, the total controller 70 conducts the following setting steps in accordance with a required recipe among a plurality of recipes provided in advance. That is to say, as described in JP Kokai No. 2000-105203, the total controller 70 first sets a wafer irradiation luminous quantity appropriate for the luminous quantity adjusting filter (luminous attenuation filter) 22, in step S296 in accordance with contamination detection sensitivity. Next, in step S297, depending on whether top layer or underlayer defects are to be detected, the total controller 70 sets the appropriate kind of polarized light to be irradiated onto the wafer via the polarizing plate 23, as described in
After this, defects are actually inspected. As shown in
When the defect inspection is completed, the total controller 70 conducts step S320 to display inspection results on a monitor 72, save inspection data in the storage unit 73, and transfer the data to a server and/or the like. After that, the total controller conducts step S321, as required, to review the contamination and other defects through the reviewing optical microscope 80, and then conducts step S322 to unload the wafer. This completes the successive inspection sequence.
Next, a more specific example of an image processor 40 is described below using
The contamination detection processors 44a-44n are associated with, for example, merge operators of 1×1, 3×3, 5×5, etc. up to n×n, and respectively include pixel-merging circuit blocks 441a-441n, 442a-442n, contamination detection processing circuits 443a-443n, and inspection region processors 444a-444n.
The present invention is characterized particularly by the contamination detection processors 44a-44n, the feature calculating circuit 45, and the integrated processor 46.
Next, the operation is described below. First, a contamination signal that has been obtained using the image sensor 36 is digitalized with the A/D converter 41. The thus-obtained detection image signal “f(i, j)” 410 is saved in the data storage unit 42 and sent to the threshold calculation processor 43. The threshold calculation processor 43 calculates a threshold image signal “Th (i, j)” 430, and in accordance with a signal that has undergone processing with the pixel-merging circuit 441, 442, the contamination detection processor 443 detects contamination for each merge operator. The inspection region processor 444 processes both the detected contamination signal and the threshold image signal with appropriate processing according to a location at which the contamination has been detected. At the same time, in accordance with the signals obtained from the pixel-merging circuit blocks 441a-441n, 442a-442n, contamination detection processing circuits 443a-443n, and inspection region processors 444a-444n of the contamination detection processors 44a-44n provided for each merge operator, the feature calculating circuit 45 calculates features such as the amount of scattered light obtained during low-angle irradiation (medium-angle irradiation included), the amount of scattered light obtained during high-angle irradiation, and the number of pixels detected on the defect. The above contamination signal and the above features are integrated by the integrated processor 46, and inspection results are displayed on the result display unit 72.
The details are described below. The A/D converter 41 is a circuit having a function that converts the analog signal that has been obtained by the image sensor 36, into a digital signal. The number of bits into which the analog signal is to be converted is desirably from about 8 to about 12. This is because there are the disadvantages that whereas conversion into a smaller number of bits results in lower resolution of signal processing and thus makes it more difficult to detect weak light, conversion into a larger number of bits makes the A/D converter more expensive, resulting in increased device price. The data storage unit 42 is a circuit for storage of the digital signal that has been obtained from the A/D conversion. Next, the threshold calculation processor 43 will be described using
Next, the operation is described below. The detection image signal “f(i, j)” that has been obtained by the A/D converter 41 is input to the delay circuit block (delay memory block) 43a and the differential processing circuit block 43b, in which the input signal is then converted into the differential signal “ΔS(i, j)”. For chip comparison, “(i, j)” denotes internal pixel coordinates of the chips. The delay circuit block 43a creates the reference image signal “g(i, j)” by delaying the input signal according to a particular iteration pitch of a circuit pattern present on the substrate under inspection. The delay circuit block 43a is only required to be a circuit that delays signal data for on-wafer chip. The amount of the signal data processed in this circuit block, however, does not always need to be equivalent to data for one chip, and the delay circuit block may function properly according to inspection target. That is to say, for a wafer having an iterative pattern in units of multiple chips, the delay circuit block may be required only to delay signal data for the multiple chips. Alternatively, for a DRAM product with an iterative pattern in chips, the delay circuit block may be required only to delay signal data according to the particular amount of iteration. The amount of signal delays can be modified under CPU instructions or the like.
The differential signal “ΔS(i, j)” that has been processed in the differential processing circuit block 43b is next stored into the data storage unit 431. The squared data calculator 43d calculates a squared value (ΔS2) of the differential signal “ΔS”, and the squared data sum calculator 43e calculates a squared data sum (ΣΔS2) of the differential signal “ΔS”. The sum calculator 43f calculates a sum (ΣΔS) of the differential signal “ΔS”. The counter 43g calculates a value denoting the number of input signals, “n”. The maximum/minimum data remover 43c has a function that calculates maximum and minimum values from multiple sets of input data and then removes the maximum and minimum values from the multiple sets of input data. The standard deviation calculator 43h uses the above-processed calculation results to calculate a standard deviation [ρ(ΔS)=√(ΣΔS2/n−ΣΔS2/n)] of the input data, and the mean value calculator 43i calculates a mean value [μ(ΔS)=ΣΔS2/n]. Furthermore, the coefficient calculator 43j calculates coefficients (magnifications) “k” for setting the threshold levels associated with the number of sets of input data, “n”.
The threshold calculator 43h uses the above-described data to calculate the image “Th(i, j)” 430 of the threshold levels [Th(H), Th(L)]. The threshold levels [Th(H), Th(L)] are calculated using expression (1) below, where, the value that has been calculated by the mean value calculator 43j is μ(ΔS), the value that has been calculated by the standard deviation calculator 43h is ρ(ΔS), the value that has been calculated or set by the coefficient calculator 43j is “k”, and a coefficient for verification is “m” (m is smaller than 1).
Th(H)=μ+kxρor Th(L)=μ−kxρor Th(Hm)=mx(μ+kxρ) or Th(Lm)=mx(μ−kxρ) (1)
The threshold images {detection threshold levels [Th(H), Th(L)] and verification threshold levels [Th(Hm), Th(Lm)]} that have thus been calculated by the threshold calculator 43k may be stored into the data storage unit 431, from which the threshold images may then be output as threshold image data. In addition, the threshold image data may be modified for each region which has been set up from the inspection region processors 444a-444n. In short, a threshold level map can be created and stored in the data storage unit 431. Furthermore, for reduced detection sensitivity in a certain region, the threshold levels in the region may be enhanced.
Next, signal pixel-merging circuit blocks 441, 442 are described below using
Since each threshold image consists of four image signals [Th(H), Th(Hm), Th(Lm), Th(L)] as described above, four merge operators “Ops” are required in each pixel-merging circuit 442a-442n. Therefore, the detection image signal 410 is subjected to merging processing in various merge operators 1504 before being output as a merged detection image signal 4411a-4411n from each pixel-merging circuit 442a-442n. In the meantime, the four threshold image signals 430 [Th(H) (detection threshold level “a”) 430(1), Th(Hm) (detection threshold level “a”) 430(2), Th(Lm) (detection threshold level “b”) 430(3), Th(L) (detection threshold level “b”) 430(4)] are merged by merge operators “Op1”-“Opn” before being output as merged threshold image signals 4421a [4420a(1)-4420a(4) to 4420n (4420n(1)-4420n (4420n(4)] from the pixel-merging circuits 442a-442n, respectively. The merge operators in each pixel-merging circuit 442a-442n perform the same function.
A pixel-merging effect will be described here. In the defect inspection apparatus of the present invention, it is necessary to detect defects without overlooking not only microcontamination, but also large thin-film-like contamination spread over a range of several μm. However, since thin-film-like contamination does not always produce strong detection image signals, the detection image signals generated in one-pixel units may be too low in S/N ratio to prevent overlooking. If an average detection image signal level of defect sampling in one-pixel units is taken as S, and an average variation as σ/n, extracting pixels in “n×n” pixel units equivalent to a size of thin-film-like contamination and then conducting convolutional computations will result in a detection image signal level of n2×S and a variation (N) of n×σ. The S/N ratio obtained in this case, therefore, will be n×S/σ. Meanwhile, if thin-film-like contamination is detected in one-pixel units, the detection image signal level and variation obtained will be S and σ, respectively, and thus the S/N ratio obtained will be S/σ. Convolutional computation of the pixels extracted in “n×n” pixel units equivalent to the size of thin-film-like contamination, therefore, makes it possible to improve the S/N ratio by a factor of “n”.
For contamination as small as about one pixel unit, the detection image signal level detected in one-pixel units is S and a variation is σ, so the S/N ratio obtained is S/σ. If pixels are extracted in “n×n” pixel units from an image of contamination as small as about one pixel unit and convolutionally computed, a detection image signal level of S/n2 is obtained with a variation of n×σ, so the S/N ratio obtained is S/n3/σ. For contamination as small as about one pixel unit, therefore, higher S/N ratios can be obtained by acquiring a signal in a pixel unit.
In addition, while the function of the merge operators described in the present embodiment has been described in the example of output of an “n×n” pixel mean value, the maximum and minimum values of “n×n” pixels or a median value thereof may be output instead. A stable signal is obtained if the median value is used. Furthermore, the output value may be a value obtained by multiplying or dividing the mean value of “n×n” pixels by a specific value.
Next, a more specific example of the contamination detection processor 443 is described below using
Next, the inspection region processors 444a-444n are described below. Each inspection region processor 444a-444n is used to remove data of regions not requiring inspection (including in-chip region) from the contamination or defect detection signal obtained from the particular chip by the contamination detection processing circuits 443a-443n. Each of the inspection region processors 444a-444n is also used to change detection sensitivity for each region (including the in-chip region), and to select a region to be inspected. For the inspection region processors 444a-444n, if detection sensitivity is allowed to be reduced for a specific region on the substrate 1 to be inspected, the threshold levels for the region, obtained from the threshold calculator 43k of the threshold calculation processor 43, may be set to be higher, or among all contamination data output from the contamination detection processing circuits 443a-443n, only data of the contamination existing in regions to be inspected may be left, based on the coordinates of the contamination.
The region for which the detection sensitivity is allowed to be reduced refers to a region of low circuit pattern density on the inspection target substrate 1. An advantage obtained from reducing the detection sensitivity is that a detection quantity can be reduced efficiently. That is to say, a highly sensitive inspection apparatus may need to detect several tens of thousands of foreign substances. If this is the case, contamination that may exist in the regions having a circuit pattern(s) is of major interest, and taking the proper measures against the contamination of major interest is a shortcut leading to improvement of device manufacture in yield. If all regions on the inspection target substrate 1 are inspected with the same sensitivity, however, this causes the contamination of major interest and contamination of minor interest to be mixed, thus impeding easy extraction of the former contamination. For these reasons, on the basis of in-chip CAD information or threshold level map information, the inspection region processors 444a-444n can extract the contamination of major interest efficiently by reducing the detection sensitivity of circuit pattern-free regions which are not so influenced on yield. A method of extracting the contamination is not limited to changing the detection sensitivity. An alternative extraction method for the contamination of major interest may be based on the classification of contamination that will be described later herein, or based on contamination sizes.
Next, the integrated processor 46 and its inspection result display unit 72 are described below. The integrated processor 46 has functions that integrate the contamination detection results that were concurrently processed by the pixel-merging circuits 441, 442, integrate the contamination detection results and the feature data that was calculated by the feature calculating circuit 45, and send the contamination detection results and the feature calculation results to the result display unit 72. This inspection result integration process is desirably performed using, for example, a personal computer (PC) to facilitate to change contents of the process.
First, the feature calculating circuit 45 is described below. The features here mean the values that represent features of the detected contamination or defects, and the feature calculating circuit 45 is a processing circuit that calculates the features. Examples of the features include the amounts of reflected/diffracted light (scattered light) (Dh, DI) from contamination or defects, obtained during high-angle illumination/overhead detection, low-angle illumination/overhead detection, and low-angle illumination/oblique detection, the number of detected pixels, shapes of the contamination detection regions, directions of principal axes of inertia, the detection location of the contamination on the wafer, the kind of underlayer circuit pattern, the detection threshold levels of the contamination, and so on.
Next, an example of darkfield classification (DFC) by the integrated processor 46 is described below. Since the contamination detection signals that underwent various pixel-merging processes are input to the integrated processor 46, this processor can classify contamination into “Large contamination”, “Microcontamination”, and “Contamination small in height”, as shown in
Next, an example of a contamination size measuring method according to the present invention is described below. This method utilizes the fact that there is a proportional relationship between the contamination size and the luminous quantity detected by the image sensor 36. That is to say, both are maintained in the relationship that for small contamination, in particular, the luminous quantity detected, D, is proportional to the sixth power of contamination size G, subject to Mie scattering theory. In accordance with the detected luminous quantity D, the contamination size G, and a proportional coefficient “ε”, the feature calculating circuit 45 can measure the contamination size using expression (2) shown below, and supply measurement results to the integrated processor 46.
G=ε×D(⅙) (2)
A value that has been calculated in advance from the luminous quantity detected from contamination of a known size is input as the proportional coefficient “ε”.
The luminous quantity D detected from microcontamination is calculated by summing up pixel values of the contamination signal section. Calculation accuracy of the detected luminous quantity D can be improved by correcting a saturated portion of the contamination signal section by Gaussian distribution approximation. Although the present example uses the value of the signal sum of the contamination signal section as the detected luminous quantity, this value does not always need to be the signal sum and may be a maximum value of the contamination signal section instead. Using the maximum value yields an advantage in that an associated electrical circuit scale can be made smaller, and using the signal sum produces an advantage in that a signal-sampling error rate can be reduced for more stable results.
Next, another example of contamination or defect classification by the integrated processor 46 is described below using
First, the wafer 1 is inspected using first inspection parameters in step S221. The coordinate data on various kinds of contamination, obtained from the contamination detection processing circuit 443 during the first inspection, and the features of each kind of contamination, obtained from the feature calculating circuit 45 during the first inspection, are saved in a storage unit (not shown) in step S222.
Next, in step S223, the wafer 1 is inspected using second inspection parameters different from the first ones. The coordinate data on various kinds of contamination, obtained from the contamination detection processing circuit 443 during the second inspection, and the features of each kind of contamination, obtained from the feature calculating circuit 45 during the second inspection, are saved in the storage unit (not shown) in step S222. For example, if the first inspection parameters are used for low-angle illumination, that is, for irradiating the wafer with illumination light from an angle close to the wafer surface, high-angle illumination parameters for irradiating the wafer with illumination light from an angle close to a normal line to the wafer surface are preferably selected as the second inspection parameters. In addition, when the wafer 1 is inspected using the second inspection parameters, the features at the coordinates where the contamination has been detected with the first inspection parameters are stored, regardless of whether some contamination is detected with the second inspection parameters.
Coordinate data that has been obtained from the first inspection results, and the coordinate data that has been obtained from the second inspection results are next compared in step S225. In step S226, if the two sets of coordinates are close to each other, the two pieces of contamination are regarded as identical contamination, and classified from the respective features. Here is an example of a method of judging the two sets of coordinate data to be close to each other: if the coordinate data obtained from the first inspection results is expressed as x1 and y1, and the coordinate data obtained from the second inspection results, as x2 and y2, and a comparison radius as “r”, data falling under expression (3) below can be determined to be identical.
(x1−x2)2+(y1−y2)2<r2 (3)
where, “r” can be 0 or a value that allows for an error component inherent in the apparatus. This value can be measured by, for example, calculating a value of the left side of expression (3) for several sampling points of contamination coordinate data, then assigning associated mean value and standard deviation to expression (4), and assigning a value derived from expression (4), as “r”.
r2=Mean value+3×Standard deviation (4)
While the present example has been described assuming that two inspecting operations are conducted, if increasing the number of kinds of features (such as a detected-pixel count equivalent to area Q of a defect) is likely to be effective for improving classification performance, inspection may be repeated at least three times to acquire the features of various kinds of contamination (detected-pixel counts).
Next, yet another example of contamination or defect classification by the inspection results integrated processor 46 is described below using
Hereunder, concepts of classification methods are described using
Classification boundaries (for total classification categories—1) are set in the features space. In
The defects can be classified into at least three categories (e.g., contamination defects, scratch defects, and circuit pattern defects) particularly by acquiring the amount of light scattered (Dh)] from the defect during high-angle illumination, the amount of light scattered (DI)] from the defect during low-angle illumination, and the number of pixels detected in the defect during the high-angle illumination and the low-angle illumination, as the three kinds of features. In this case, since the number of pixels detected in the defect (planar area Q of the defect) is acquired as a feature, contamination defects can be further categorized into large contamination and small contamination.
In addition, at least the sub-categorization of the contamination defects into large contamination defects and small ones becomes easy by acquiring three features, namely, the amount of light scattered from the defect at a high imaging magnification, the amount of light scattered from the defect at a low imaging magnification, and the number of pixels detected in the defect. Furthermore, classification of any microcontamination, scratches, and other defects that may exist on a transparent film becomes possible by acquiring features of a defect image obtained from a photodetector 640.
By the way,
It is also effective to display an isolation ratio, since a user can easily understand isolation performance. The isolation ratio here means, for example, a rate of the contamination that falls under one category and that is included in the regions isolated by an isolation boundary.
While the present example has been described assuming the use of the features calculated using three kinds of optical parameters, the number of kinds of optical parameters used does not always need to be limited to three and the present invention is applicable in cases where features can be calculated using multiple kinds of optical parameters, and where a plurality of features can be acquired using multiple kinds of optical parameters.
Next, a description is given below of another example in which the inspection results obtained from the image processor 40 are displayed on, for example, the display unit 72 by the total controller 70.
Alternatively, positions of the detected substances displayed as the review images 3903 may also be displayed on the inspection map 3901, or detection numbers of the detected substances may also be displayed on the review images 3903. In addition, while the present embodiment has been described assuming the display of six review images per set, the display format does not need to be limited to the display of six review images per set; all detected contamination or defects may be displayed or only a fixed rate of substances with respect to a detection count thereof may be displayed.
In a further alternative display format, the detected substances that have been classified into two kinds such as contamination and scratches may be displayed and their classification accuracy may also be displayed together.
Next, the inspection parameter (inspection recipe) setup and other processes performed in the total controller 70 and/or the like are described below using
Next, each setting step that the total controller 70 executes is described below. In the chip layout setting step (S211), chip sizes and whether a chip is present on the wafer are set to the image processor 40 and the like in accordance with CAD information and the like. The chip sizes must be specified since they corresponds to the distances required for comparative processing. Next, rotational matching (step S212) is executed to rotate the wafer 1 for its rotational shift of about “0”. In other words, step S212 is executed to rotate the wafer 1 such that a layout direction of the chips on the wafer 1 rested on the stage which the total controller 70 causes the transport system 10 to control is adjusted to be parallel to the pixel direction in the image sensor 36. The execution of the rotational matching step arranges the iterative patterns on the wafer 1 uni-axially, hence making it possible to easily execute chip comparative signal processing. In the inspection region setting step (S213), desired inspection locations on the wafer where the inspection is to be executed, and detection sensitivity for each specified region are specified for the image processor 40 controlled by the total controller 70. Each region on the wafer can be inspected with optimal sensitivity by executing the inspection region setting step (S213).
In the optical parameter setting step (S214), an irradiation direction and irradiation angle of the wafer irradiation with the slit-shaped beam 90, 90′, whether the coherency reduction optical system 230 is to be used, and a magnification of the detection optical system 30 are selected for the irradiation optical system 20 and detection optical system 30 controlled by the total controller 70. The above selections can be conducted in such an optical parameter setting window as shown in
In the optical filter setting step (S215), whether the spatial filter 32 and polarizing filter 24b shown in
In the luminous quantity detection setting step (S216), the amount of light entering the image sensor 36 is adjusted for the irradiation optical system 20 or detection optical system 30 controlled by the total controller 70. For the light reflected/scattered from the circuit patterns formed on the wafer, the component type of the light scattered differs according to the particular shape of the pattern. More specifically, for a flat wafer surface, scattered light does not occur so much and most of the light is regularly reflected light. For a very rough wafer surface, however, a great deal of scattered light occurs. In other words, the amount of light reflected/scattered from the circuit pattern varies according to the particular state of the wafer surface, that is, the kind of device-manufacturing process. However, since the image sensor 36 has its own dynamic range, adjustments are desirably conducted such that an appropriate amount of light according to the dynamic range will enter. For example, adjustments are desirably conducted such that the amount of light reflected/scattered from the circuit pattern on the wafer will be about 1/10 of the dynamic range of the image sensor 36. The amount of light entering the image sensor 36 may be adjusted by controlling an output luminous quantity with the luminous quantity adjusting filter 22 or by conducting adjustments with the ND filter 34.
The signal-processing parameter setting step (S217) is executed to set up the contamination/defect detection parameters that the total controller 70 causes the image processor 40 to control.
The user can perform inspections with desired parameters by starting the inspection process (step S218) after the above steps have been executed.
The parameters that have been described in the present example may be set up, for example, by manually entering data based on design information of the inspection target, by entering data with an input assist function accompanying the defect inspection apparatus of the present invention, or by acquiring information from a host system via a network.
Of all data settings of the parameters in the above-described setting steps, at least data settings of the parameters in inspection region setting step S213, optical parameter setting step S214, optical filter setting step S215, detection luminous quantity setting step S216, and signal-processing parameter setting step S217 do not always require modification for the particular inspection target and may remain fixed, irrespective of the kind of inspection target. Although maintaining fixed parameter settings leads to reduced inspection parameter setting time, each parameter is desirably tuned for higher sensitivity. In addition, inspection region setting step S213 does not always need to precede optical parameter setting step S214 and may be conducted in any phase before inspection process step S218 is conducted.
An example of a display screen menu for setting up the above-described parameters is shown in
Next, details are described below. The parameter setting sequence diagram 4301 indicates a process flow of the inspection parameter setup in the defect inspection apparatus of the present invention. The user sets up parameters in order from “Chip layout setting” in accordance with the parameter setting sequence diagram.
A feature of the parameter setting sequence diagram 4301 is that since the process flow of the parameter setting is indicated by arrow 4305, the user can set each parameter most efficiently without making a mistake in the setting order. Another feature exists in that all items are divided into items that are necessary to be set (mandatory items) and items that are not necessary to be set (non-mandatory items). The latter items are those whose predefined values can be used. Since the display items are divided in this way, the user can understand the minimum setup items required. In addition, to obtain inspection results immediately, the user needs only to set up the mandatory items. Also, to tune detection sensitivity, the user needs only to set up parameters in the non-mandatory setup items. That is to say, parameter setup levels can be modified to fit particular needs of the user. For example, a button 4306 is displayed in a triple-frame format to indicate that the item is a mandatory one, and a button 4307 displayed in a single-frame format indicates that the item is low in importance of setup. Yet another feature is that the menu explicitly indicates an item currently being set up by the user. A button 4308, for example, is shaded for distinction from the buttons 4306, 4307. Explicitly indicating the current setup item in this way offers an advantage in that the user can readily confirm the number of remaining setup items.
In the present example, an optional parameter setting step 4309 is added to the sequence described using
The detailed setup parameter list 4302 is a window for setting up details of each parameter item. In order to enter or select an item, the item may have a keyboard input field as with an input box 4310, or may be of an icon-based input item selection scheme using with input icons 4311, for example. The input icons 4311 are an example of icons displayed for three input items such that a press of either icon displays another window to allow further detailed parameter setup. Additionally, a necessary item may be selected, as with an input checkbox 4312.
The settings display change button 4303 changes or customizes a display item. For example, when there is an item that the user always needs to set up for each inspection operation or when the user needs to set up more items, the user can conduct changes by using the button 4303 to customize the display into a more easily usable one for enhanced inspection parameter setup efficiency. The help information button 4304 outputs user assistance information when pressed by the user to obtain further detailed information on setup procedures or on parameter settings. Details of each setup item may be output as vocal guidance messages, or information on operations may be supplied as video images in a format such as MPEG. Yet another form of supplying the user help information may be such that the user can talk with a manufacturer's designer of the defect inspection apparatus of the present invention on-line through a network or a telephone line.
As described above, according to the present invention, very small particle-like contamination (including microcontamination of about 0.1 μm or less in size), thin-film-like contamination, scratches, and other defects present on various inspection target substrates each having various regions can be inspected with high speed and high accuracy with a simple configuration.
According to the present invention, a highly efficient manufacturing line for substrates can be constructed with a simple configuration by using the above-described apparatus as an in-line monitor with a low cost.
The invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiment is therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims, rather than by the foregoing description, and all changes that come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
Claims
1. A defect inspection apparatus comprising:
- an irradiation optical system arranged to guide a light beam emitted from a light source to a first or a second optical path, to focus the light beam that has been guided to the first or second optical path into a slit-shaped light beam at a surface of a substrate having circuit patterns as an object of inspection;
- a detection optical system arranged to receive reflected/scattered light from the substrate by the irradiation of the slit-shaped light beam and to convert the received light into a signal; and
- an image processor to process the signal for defect;
- wherein each of the first optical path and the second optical path of the irradiation optical system includes a cylindrical lens adapted to focus a plurality of slit-shaped light sub-beams into the slit-shaped light beam irradiated to the surface of the substrate, and a coherency reduction optical system adapted to receive the light beam having an optical coherency and to emit the plurality of slit-shaped light sub-beams which are spatially reduced in coherency in a light-converging direction of the cylindrical lens, and
- wherein the image processor calculates features of the signal indicative of defect based on a first inspection signal obtained by receiving light from the substrate irradiated by the slit-shaped light beam from the first optical path and a second inspection signal obtained by receiving light from the substrate irradiated by the slit-shaped light beam from the second optical path, and classifies kind of defect by using information of the calculated features.
2. The defect inspection apparatus according to claim 1, wherein the coherency reduction optical system includes an optical member group formed in the first or second optical path by stacking a plurality of plate-shaped optical members each having a different optical path length in at least the light-converging direction of the cylindrical lens.
3. The defect inspection apparatus according to claim 2, wherein the coherency reduction optical system is constructed so as to be insertable into and retractable from the first or second optical path.
4. The defect inspection apparatus according to claim 2, wherein the coherency reduction optical system further includes beam-expanding optics disposed on an incident side of the optical member group in order to expand the light beam in at least a stacking direction of the plurality of plate-shaped optical members.
5. The defect inspection apparatus according to claim 2, wherein the irradiation optical system further includes an inclined mirror for irradiating the slit-shaped beam at an angle of inclination to the surface of the substrate.
6. The defect inspection apparatus according to claim 1, wherein the slit-shaped light beam has an elongated length in a first direction and is irradiated from the oblique direction to the circuit pattern of the substrate so that the elongated length extends in a direction substantially parallel to a major straight line extension group of the circuit patterns.
7. The defect inspection apparatus according to claim 1, wherein the coherency reduction optical system includes an optical member group formed by stacking a plurality of plate-shaped optical members each having a different optical path length in at least the light-converging direction of the cylindrical lens.
8. The defect inspection apparatus according to claim 7, wherein the coherency reduction optical system is constructed so as to be insertable into and retractable from the first or second optical path.
9. The defect inspection apparatus according to claim 1, wherein the detection optical system further includes a spatial filter for shielding an interference pattern formed by reflected light of repeated circuit patterns of the substrate.
10. The defect inspection apparatus according to claim 1, wherein the irradiation optical system further includes a luminous quantity adjusting filter for adjusting the amount of light which is disposed in the first or second optical path.
11. The defect inspection apparatus according to claim 1, wherein the irradiation optical system further includes a polarizing plate for conducting polarization control which is disposed in the first or second optical path, and the detection optical system further includes an analyzer that controls and detects polarization.
12. The defect inspection apparatus according to claim 1, wherein the laser light source is constructed so as to emit at least one of an UV laser light, a FUV laser light, a DUV laser light, a vacuum UV laser light, an argon laser light, a nitrogen laser light, a helium-cadmium laser light, an excimer laser light, and a semiconductor laser light.
13. A defect inspection apparatus comprising:
- an irradiation optical system arranged to guide a light beam emitted from a light source to a first or a second optical path, to focus the light beam that has been guided to the first or second optical path into a slit-shaped light beam at a surface of a substrate having circuit patterns as an object of inspection with the same elevation angle but different azimuth angles;
- a detection optical system arranged to receive reflected/scattered light from the substrate by the irradiation of the slit-shaped light beam and convert the received light into an image signal; and
- an image processor arranged to process the image signal and extract a signal indicative of a defect,
- wherein the image processor calculates features of the extracted signal indicative of the defect based on a first inspection signal obtained by receiving light from the substrate irradiated by the slit-shaped light beam from the first optical path and a second inspection signal obtained by receiving light from the substrate irradiated by the slit-shaped light beam from the second optical path, and then classifies types of defects by using information of the calculated features based on the first inspection signal and the second inspection signal.
14. The defect inspection apparatus according to claim 13, wherein each of the first optical path and the second optical path of the irradiation optical system includes a cylindrical lens and a coherency reduction optical system, the coherency reduction optical system being adapted to receive the light beam having optical coherency and to emit a plurality of slit-shaped light sub-beams which are spatially reduced in coherency in a light-converging direction of the cylindrical lens, the cylindrical lens being adapted to focus the plurality of slit-shaped light sub-beams into the slit-shaped light beam irradiated to the surface of the substrate.
15. The defect inspection apparatus according to claim 14, wherein the coherency reduction optical system includes:
- an optical member group formed in the first or second optical path by stacking a plurality of plate-shaped optical members each having a different optical path length in at least the light-converging direction of the cylindrical lens;
- beam-expanding optics disposed on an incident side of the optical member group in order to expand the light beam in at least a stacking direction of the plurality of plate-shaped optical members; and
- an inclined mirror for irradiating the slit-shaped beam at an angle of inclination to the surface of the substrate.
16. The defect inspection apparatus according to claim 14, wherein the slit-shaped light beam has an elongated length in a first direction and is irradiated from the oblique direction to the circuit pattern of the substrate so that the elongated length extends in a direction substantially parallel to a major straight line extension group of the circuit patterns.
17. The defect inspection apparatus according to claim 14, wherein the detection optical system further includes a spatial filter for shielding an interference pattern formed by reflected light of repeated circuit patterns of the substrate.
18. The defect inspection apparatus according to claim 14, wherein the irradiation optical system further includes:
- a luminous quantity adjusting filter for adjusting the amount of light which is disposed in the principal optical path; and
- a polarizing plate for conducting polarization control which is disposed in the principal optical path.
19. The defect inspection apparatus according to claim 14, wherein the laser light source is constructed so as to emit at least one of an UV laser light, a FUV laser light, a DUV laser light, a vacuum UV laser light, an argon laser light, a nitrogen laser light, a helium-cadmium laser light, an excimer laser light, and a semiconductor laser light.
20. The defect inspection apparatus according to claim 14, wherein the coherency reduction optical system is constructed so as to be insertable into and retractable from the first or second optical path.
Type: Application
Filed: Jun 30, 2010
Publication Date: Oct 21, 2010
Inventors: Hiroyuki Nakano (Chigasaki), Akira Hamamatsu (Yokohama), Sachio Uto (Yokohama), Yoshimasa Oshima (Yokohama), Hidetoshi Nishiyama (Hitachinaka), Yuta Urano (Yokohama), Shunji Maeda (Yokohama)
Application Number: 12/827,470
International Classification: G01N 21/00 (20060101); G06K 9/00 (20060101);
