This application is based upon and claims the benefit of priority from the prior Japanese Patent Application No. 2011-068485, filed on Mar. 25, 2011, the entire contents of which are Incorporated herein by reference.
Embodiments described herein relate generally to a pattern inspection apparatus and a pattern inspection method.
In the fields of semiconductor devices, flat panel displays, and micro electro mechanical systems (MEMS), a structure (hereinafter referred to as a “microstructure”) having a micropattern formed on its surface is manufactured by the use of, for example, a lithography technique.
An optical inspection apparatus is used for the inspection of such a microstructure. The conventional inspection apparatus applies a light to an inspection target pattern formed on a substrate such as a wafer from, for example, a laser light source or a lamp light source, detects a reflected light from the pattern by a detector, and compares the signal intensity of the detected light by, for example, die-to-die comparison, thereby conducting a defect inspection.
Recently, patterns having high aspect ratios have been produced due to advanced miniaturization and integration of the microstructures. For example, in a trench pattern having a high aspect ratio, defects can be generated at various locations in a depth direction (see
However, there have heretofore been a problem of defect detection failures resulting from a small amount of information only obtained by the signal intensity of the detected light, and a problem of the decrease of the S/N ratio of a defect dependent on the height (depth) of the defect.
In the accompanying drawings:
In accordance with an embodiment, a pattern inspection method includes: applying a light generated from a light source to the same region of a substrate in which an inspection target pattern is formed; guiding, imaging and then detecting a reflected light from the substrate, and acquiring a detection signal for each of a plurality of different wavelengths; and adding the detection signals of the different wavelengths in association with an Incident position of an imaging surface to generate added image data including information on a wavelength and signal intensity, judging, by the added image data, whether the inspection target pattern has any defect, and when judging that the inspection target pattern has a defect, detecting the position of the defect in a direction perpendicular to the substrate. Embodiments will now be explained with reference to the accompanying drawings. Throughout the drawings, like components are given like reference signs, and detailed explanations of such components are omitted when not needed.
A comparative example examined by the present inventor in the process of making the invention is illustrated.
As shown in the flowchart of
Recently, the aspect ratio of an inspection target pattern has been increasingly high due to more advanced miniaturization and integration of microstructures. For example, there is a trench pattern having an aspect ratio of more than 40. In such a pattern having a high aspect ratio, defects of various depths can be formed (see reference signs DF1 and DF2 in
Now, the pattern inspection apparatus according to the first embodiment Is described.
First, the principle of pattern inspection on which the present embodiment is based is roughly described.
As apparent from
The relation between the heights (depths) and wavelengths of the defects is prepared as a data table to conduct pattern inspections using lights of a plurality of wavelengths, and information on the wavelengths and signal intensity of the detected defect candidates is contrasted with the data table pixel by pixel, that is, in association with the incident positions of a detection surface. This makes it possible to detect not only the position of a defect in a two-dimensional plane but also its height (depth).
A pattern inspection apparatus 1 shown in
The illumination optical system 10 includes a white light source 12 and a spectroscope 18. The white light source 12 emits a broad light. The spectroscope 18 disperses the broad light into lights of a plurality of different desired wavelengths.
The Illumination unit to apply lights of a plurality of different wavelengths to the wafer W is not limited to the white light source 12 and the spectroscope 18. For example, a plurality of LED light sources or laser light sources that emit lights of different wavelengths may be used, and the lights may be switched in the order of wavelengths by a light source switching device and applied accordingly.
For example, a substrate similar to the filter board shown in
Returning to
The imaging optical system 30 includes an objective lens 32 and an imaging lens 34. The Imaging optical system 30 controls the optical path of the reflected light to form an optical image of the pattern P1 and Its periphery on the detection surface of a detector 50.
The detector 50 photoelectrically converts the reflected light Imaged on the detection surface, and outputs a detection signal to the image processor 60. The detector 60 comprises, for example, an infrared charge coupled device (CCD) or a photomultiplier. However, the detector is not limited to such devices. Any device that can photoelectrically convert the imaged light can be suitably selected.
The image processor 60 includes a signal processor 62, a calculator 64, a defect candidate judge 66, and a height (depth) output 68. The detection signal is provided to the signal processor 62 from the detector 50, and the signal processor 62 creates image data on the wavelength and signal intensity in association with each pixel of the detector 50. The data is provided to the calculator 64 from the signal processor 62, and the calculator 64 adds the data pixel by pixel and creates added image data on the wavelength and signal intensity associated with each pixel position. This added image data is, for example, three-dimensional data in which wavelength information is added in a z-direction to signal intensity information described in addition to (x, y) indicating the coordinate position of a pixel. In this case, the z-direction is the wavelength of the dispersed light.
The three-dimensional data thus created is stored in a memory MR1. Three-dimensional data obtained for a reference pattern is also stored in the memory MR1.
The defect candidate judge 66 compares the three-dimensional data created by the calculator 64 for the inspection target pattern P1 with the three-dimensional data for the reference pattern taken from the memory MR1, thereby judging whether the pattern P1 has any defect. More specifically, when there is data different from the three-dimensional data for the reference pattern, the defect candidate judge 66 judges that there is a defect candidate at the corresponding pixel position, that is, at a position on the wafer W that corresponds to the position at which the reflected light enters the detection surface of the detector 50.
A pattern determined to have no defect may be used as the reference pattern. However, in the present embodiment, a pattern P2 formed in an adjacent cell or die to have the same shape and dimensions as the pattern P1 is used for a defect judgment by die-to-die comparison. In the present embodiment, the pattern P2 corresponds to, for example, a second pattern.
Three-dimensional data for a defect candidate is provided to the height (depth) output 68 from the defect candidate judge 66, and the height (depth) output 68 judges the height or depth of the defect candidate by reference to the data table described above with reference to
The monitor 80 displays, for example, by a liquid crystal display, the position and height (depth) of the defect candidate judged by the image processor 60.
Here, a defect candidate judged by the defect candidate judge 66 may have, depending on its wavelength, signal intensity lower than those of other wavelengths. In this case, the defect candidate judge 66 adds a predetermined offset to the signal intensity of a desired wavelength. As a result, the defect candidate is displayed on the monitor 80 in an enhanced form.
The stage S allows the wafer W to be mounted thereon, and moves the wafer W in a direction level with the waver surface in accordance with a control signal provided from the stage controller 90. Thus, the wafer W is scanned with the illumination light in the direction level with the waver surface. The stage controller 90 generates a control signal for driving the stage 40 in accordance with a command signal from the controller 30.
The controller 40 generate various control signals, and sends the control signals to the illumination optical system 10, the stage controller 90, the detector 50, and the image processor 60. The image data is sent to the controller 40 from the image processor 60, and the controller 40 detects the sensitivity of the detector 50 for each wavelength. When judging that the sensitivity is insufficient, the controller 40 generates a control signal for correcting the sensitivity, and sends this control signal to the detector 50. This makes it possible to improve the sensitivity to, for example, a reflection signal of the wavelength λ1 shown in
Now, the pattern inspection method according to the first embodiment is described.
First, lights of a plurality of different wavelengths are generated, and are applied to the same region of the wafer, for example, a region where the inspection target pattern P1 is formed (step S1). The lights of a plurality of different wavelengths can be generated by, placing, on the optical path of a white light, a plurality of wavelength filters for dispersing a light into a plurality of desired wavelengths, emitting a white light from a single light source, and properly and selectively switching the wavelength filters to disperse the white light. Alternatively, a plurality of LED light sources or laser light sources that emit lights of different wavelengths can be used, and the light sources can be switched, for example, in the order of wavelengths to generate and apply lights of a plurality of different wavelengths. The lights of a plurality of different wavelengths can also be generated by turning on a signal laser light source that emits lights of different wavelengths.
Next, the reflected light from the wafer is imaged on the detection surface of the detector, and a signal of each wavelength is acquired (step S2).
Furthermore, the acquired signal Is added, and three-dimensional data including information on the coordinate position (x, y) of the pixel, the signal intensity, and the wavelength is created (step S3).
The wafer is then moved to bring, into a field of view, the region, which is different from the region where the pattern P1 is formed, of the pattern P2 formed by a pattern equal in design to the pattern P1. In accordance with a procedure similar to that described above, three-dimensional data including information on the coordinate position (x, y) of the pixel, the signal intensity, and the wavelength for the pattern P2 is created. Whether the pattern P1 has any defect is detected by the comparison with the three-dimensional data, that is, by die-to-die comparison (step S4). The comparison with the three-dimensional data is not limited to the die-to-die comparison. For example, a reference pattern which has been determined in advance to have no defect may be used, and comparison may be made with three-dimensional data obtained for the reference pattern.
Finally, when the pattern P1 is judged to be a defect candidate, information on the height or depth of the defect candidate is output by reference to the prepared data table that shows the relation between the wavelength, the signal intensity, and the defect height (depth) (step S5).
Thus, according to the present embodiment, three-dimensional data including information on the coordinate position (x, y) of the pixel, the signal intensity, and the wavelength is created, so that the information amount is much greater than in the comparative example, and defect detection failures can therefore be inhibited. Moreover, reference is made to the data table that shows the relation between the wavelength, the signal intensity, and the defect height (depth). Therefore, it is possible to acquire not only information on the presence of a defect but also information on the height or depth of the defect, so that the S/N ratio of the defect advantageously improves independent of the height (depth) of the defect.
Now, a pattern inspection apparatus according to the second embodiment is shown by way of example with reference to
As obvious from the contrast with
According to this configuration, detection signals of a plurality of different wavelengths can also be acquired by imaging a reflected light from a wafer W on the detection surface of the color CCD camera 70.
The configuration of the pattern inspection apparatus 3 shown in
Thus, as a means of acquiring detection signals of a plurality of different wavelengths without dispersing an incident light, a prism is provided, for example, between an objective lens 32 and an imaging lens 34 in an imaging optical system 30 to divide an optical path into a plurality of optical paths, and a desired wavelength filter is disposed for each of the divided optical paths. In this way, detection signals of a plurality of different wavelengths can also be acquired by a detector similar to the detector 50 in
As another means of acquiring detection signals of a plurality of different wavelengths, a spectroscope may be configured by an actuator and by a filter board in which wavelength filters of desired wavelengths are arranged on its periphery in the same manner as the spectroscope 18 shown in
Now, a pattern inspection method according to the second embodiment is described.
First, a white light is generated from a light source, and applied to the same region of the wafer, for example, a region where an inspection target pattern P1 is formed (step S11).
Next, the reflected light from the wafer is imaged on the detection surface of the detector, and a signal of each wavelength is acquired (step S12).
In order to acquire signals of a plurality of different wavelengths, a color CCD camera may be used as a detector, or a prism for dispersing the reflected light may be provided in the imaging optical system 30 and a plurality of desired wavelength filters may be disposed on the optical path of each dispersed light. Alternatively, in order to acquire signals of a plurality of different wavelengths, a filter board having a wavelength filter of a desired wavelength disposed therein may be disposed, for example, between the objective lens 32 and the Imaging lens 34, and the board FB may be rotated to locate the wavelength filter of a desired wavelength on the optical path of the reflected light so that the light of the desired wavelength enters the detection surface of the detector
Furthermore, the acquired signal is added, and three-dimensional data including information on the coordinate position (x, y) of the pixel, the signal intensity, and the wavelength is created (step S13).
The wafer is then moved to bring, into a field of view, the region, which is different from the region where the pattern P1 is formed, of a pattern P2 formed by a pattern equal in design to the pattern P1. In accordance with a procedure similar to that described above, three-dimensional data for the pattern P2 Including information on the coordinate position (x, y) of the pixel, the signal intensity, and the wavelength is created. Whether the pattern P1 has any defect is detected by the comparison with the created three-dimensional data for the pattern P2, that is, by die-to-die comparison (step S14). The comparison with the three-dimensional data is not limited to the die-to-die comparison. For example, a reference pattern which has been determined in advance to have no defect may be used, and comparison may be made with three-dimensional data obtained for the reference pattern.
Finally, when the pattern P1 is judged to be a defect candidate, information on the height or depth of the defect candidate Is output by reference to the prepared data table that shows the relation between the wavelength, the signal intensity, and the defect height (depth) (step S15).
Thus, according to the present embodiment as well, three-dimensional data including information on the coordinate position (x, y) of the pixel, the signal intensity, and the wavelength is created, so that the Information amount is much greater than in the comparative example, and defect detection failures can therefore be inhibited. Moreover, reference is made to the data table that shows the relation between the wavelength, the signal intensity, and the defect height (depth). Therefore, it is possible to acquire not only information on the presence of a defect but also information on the height or depth of the defect, so that the S/N ratio of the defect advantageously Improves independent of the height (depth) of the defect.
In a pattern inspection targeted at a pattern formed by a thin film, light interference caused by the thickness variation of thin films results in noise. To avoid this, the light from the light source of the pattern inspection apparatus preferably has a wavelength width that can cancel the thickness variation. More specifically, a light source having a wavelength width of ±20 nm or more is preferable. To this end, it is possible to use, for example, a pulse laser of triple harmonic by Ti: sapphire and of a femtosecond order (10 to 15) having a wavelength of 260 nm ±20 nm.
Moreover, a broadband light source which is a combination of lasers of a plurality of different wavelengths can be used Instead of the pulse laser light source.
A basic light source unit 620 shown in
The relation between the wavelength and the wavelength width is:
Δλ=Δλ266 nm×(λ266 nm×/λ1064 nm)2.
Therefore, the deep ultraviolet light output from the SHG element 624b has a wavelength width of about 266 nm±1.25 pm.
A broadband light source 600 shown in
A broadband light source 700 shown in
In the embodiments described above, not only the single light source but also a plurality of light sources may be used. In this case as well, the above-mentioned pulse laser light source or broadband light source is applicable to each of the above-mentioned light sources.
While certain embodiments have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions. Indeed, the novel methods and systems described herein may be embodied in a variety of other forms; furthermore, various omissions, substitutions and changes in the form of the methods and systems described herein may be made without departing from the spirit of the inventions. The accompanying claims and their equivalents are intended to cover such forms or modifications as would fall within the scope and spirit of the inventions.
Number | Date | Country | Kind |
---|---|---|---|
2011-068485 | Mar 2011 | JP | national |