Lithography Optics Adjustment and Monitoring

Abstract

To provide methods for reducing the downtime to address the problem that current downtime is too long.SOLUTION: Provided herein are methods and apparatus for processing an image of a beam generated by an optical system to extract information indicative of an extent of damage to optical elements in the optical system. Also disclosed is a beam image and analysis tool capable of acquiring an image of a beam at any one of a number of locations.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application is a continuation-in-part of U.S. patent application Ser. No. 15 / 208,152, filed Jul. 12, 2016, the specification of which is incorporated by reference in its entirety into this specification.

[0002] The subject matter of this disclosure relates to optical components in laser-generated light sources such as those used in integrated circuit photolithography manufacturing processes. [Background technology]

[0003]

[0003] Optical elements used in light sources for semiconductor photolithography degrade over time, resulting in degradation of light source performance that cannot be repaired while the system is in operation. For example, calcium fluoride (CaF2) is a crystalline material commonly used in deep ultraviolet (DUV) light sources for a variety of applications, such as windows, beam splitters, and reflectors. Although CaF2 is a robust material for long-term use in lithography systems, degradation of CaF2 can occur through any one of several mechanisms. Aside from catastrophic failure, these mechanisms tend to gradually degrade the performance of CaF2 optical components. Initially, such degradation may be relatively insignificant, but eventually the optical performance of the affected components will degrade to the extent that the light source must be taken off-line and the optical component or a module containing the optical component replaced.

[0004]

[0004] One particular optical degradation mechanism involves the formation of slip planes. Slip planes occur when thermomechanical stresses induce small fractures in the CaF2 crystal lattice. Visually, slip planes appear as lines on the optic, positioned at distinct angles relative to the orientation of the crystal lattice. Visual inspection is only practical after removing the damaged optic from the light source, which can only occur if the light source is removed from the product.

[0005]

[0005] Another optical degradation mechanism is the formation of line damage, one form of which results from the overlap of multiple parts of the beam on an optical component having a coating. The combined energy of these multiple parts can exceed the damage threshold of the coating on the optical component. When this occurs, it can damage the coating in the overlapping area, which creates a distinct line that can diffract the light. Line damage is problematic because it can cause distortions in the beam profile that reduce the lifetime of the module. It is beneficial to be able to detect the presence and extent of line damage.

[0006]

[0006] Yet another aspect of the optical degradation mechanism is the presence of point-like defects due to particles on optical components at some location within the system. Such particles are so small in number that they usually do not have an excessively adverse effect on optical performance, but they may affect image processing. It is beneficial to be able to identify the presence, location, and nominal size of point-like defects.

[0007]

[0007] Yet another aspect of the optical degradation mechanism is the presence of worm-like features in the beam image, associated with coating damage on the OPuS beam splitter, which degrades the optical performance. It would be beneficial to be able to identify the presence, location, and nominal size of the worm-like features.

[0008]

[0008] A further damage mechanism that reduces optical performance is the occurrence and growth of dendritic structures. These dendritic structures resemble white fluff, as they are commonly referred to. The presence of dendritic structures indicates a coating failure, which leads to localized heating of the optical component in the vicinity of the dendritic structure. This heating exacerbates the coating failure, so that the affected area in the optical component expands, which further exacerbates the localized heating and causes thermal lensing. The optical performance of the laser then deteriorates, and eventually the affected optical component must be replaced. It is beneficial to be able to detect the occurrence and development of dendritic structures.

[0009]

[0009] A common aspect of the challenges associated with these damage mechanisms is the need for methods to detect and identify the low level signals associated with these defects, and to attribute these signals to specific defect types. A further need exists for methods to monitor the development of these defects that can predict how much time remains before the optical performance of the light source is degraded to the point that an optical component or module must be replaced.

[0010]

[0010] One method of detecting changes in an optical component involves imaging a portion of a laser beam that has passed through the component. This method typically involves acquiring both near-field and far-field images of the beam at the system's output. While such methods are effective in detecting optical damage, they generally cannot pinpoint where in the optical path the damage occurred. It would be beneficial to have a system that can acquire images of the beam at other locations in addition to the output.

[0011]

[0011] Eventually the light source must be removed from productive use so that degraded modules and optics can be replaced. This obviously results in downtime for the scanner. It is desirable to minimize downtime to the greatest extent possible. Although the negative impact of lost uptime can be reduced by planning maintenance events, every downtime represents a significant cost to the chip manufacturer. The amount of time that the light source is out of service is sometimes referred to as the "green-to-green" time.

[0012]

[0012] During maintenance, once the required replacement module and optics are installed, the light source needs to be realigned. As mentioned above, the light source can be embodied as a set of modules. In such cases, alignment typically involves adjusting positioners that point and / or tilt, in some cases, individual optical elements within the module, and in other cases, the entire module. Once aligned, the light source can be used for a significant amount of time without making additional adjustments. The time required to realign the optics and modules adds up unfavorably to the green-to-green time. Alignment can be performed by using mechanical alignment actuators, but this can be very time-consuming.

[0013]

[0013] Part of the alignment process may be to compensate for chamber resonances. Laser radiation for semiconductor photolithography is usually delivered as a series of pulses at a particular repetition rate. Chamber resonances may occur at some repetition rates and may cause a sharp increase in performance metrics (e.g., pointing and divergence) near the resonance frequency, with a low frequency valley or floor near the resonance point. The presence of resonances may require additional time and effort during alignment to keep the performance metrics within specifications, but may be acceptable if all data points remain within specifications. Furthermore, resonances that cause peak-to-valley differences in performance metrics may create technical challenges for scanner design and control.

[0014]

[0014] Measures to minimize the total green-to-green time include both reducing the number of maintenance events and shortening the length of maintenance events when they occur. The number of maintenance events can be reduced by extending the operational life of the modules and optics and by using reliable methods to determine the amount of damage to the optics in the field. The length of maintenance events can be reduced by reducing the time required to align replacement modules and optics. Summary of the Invention

[0015]

[0015] The following presents a simplified summary of one or more embodiments in order to provide a basic understanding of the invention. This summary is not an exhaustive overview of all contemplated embodiments, and is not intended to identify key or critical elements of all embodiments or to delineate the scope of any or all embodiments. Its sole purpose is to present some concepts of one or more embodiments in a simplified form as a prelude to the more detailed description that is presented later.

[0016]

[0016] According to one aspect, the subject matter of the present disclosure addresses the problem of current downtime being too long by providing a method to reduce downtime. The subject matter of the present disclosure also addresses the problem of extending the life of modules and optics. The subject matter of the present disclosure also addresses improving scanner performance at repetition rates at or near chamber resonances.

[0017]

[0017] The subject matter of the present disclosure addresses these issues at least in part by using electrically actuated and controlled alignment actuators to adjust the optical elements and modules instead of purely mechanical alignment actuators. The subject matter of the present disclosure also addresses these issues at least in part through the provision of an optical sensor, such as a camera, to image a fluorescent alignment screen that indicates the state of alignment. The subject matter of the present disclosure addresses these issues at least in part by using data from a beam imaging device, such as a combined autoshutter metrology module ("CASMM"), to guide the alignment. This alignment can be performed in any one of several modes of operation. For example, as described in more detail below, the alignment can be performed in a manual mode, a semi-automatic mode, a fully automatic mode, a light source managed mode, or a scanner managed mode.

[0018]

[0018] According to another aspect, providing an electrically adjustable actuator with an on-board controller provides the ability to dynamically align the light source, adjusting and verifying during pauses in active use. This also provides the ability to flatten (i.e., reduce the difference between maximum and minimum) continuous performance metrics. For example, if the scanner dictates operation at a particular repetition rate change that is near a chamber resonance frequency, the light source may adjust one or more optical components to compensate for the resonance by reducing the resonance peak.

[0019]

[0019] According to another aspect, the subject matter of the present disclosure addresses the problem of detecting the presence of one or more slip planes from a near-field image, where the detection may be automatic, and where the evolution of the slip planes may be sensed and quantified as a function of time. Here and elsewhere, "time" may refer to a chronological time measured, for example, in terms of days, weeks, or months, or an operational time measured, for example, by cumulative system uptime, pulse count, or the like, or some hybrid time determined as a function of both chronological time and operational time.

[0020]

[0020] According to another aspect, the subject matter of the present disclosure addresses determining whether a laser beam is clipping an aperture within a light source by detecting diffraction that occurs when an at least partially coherent laser beam diffracts when it hits an obstacle such as the edge of the aperture.

[0021]

[0021] According to another aspect, the subject matter of the present disclosure deals with sensing the presence of a line damage.

[0022]

[0022] More generally, according to another aspect, the subject matter of this disclosure deals with detecting the presence of a line embedded within a noisy background signal.

[0023]

[0023] According to another aspect, the subject matter of the present disclosure deals with sensing the presence of a point-like defect, such as a particle, on an optical component.

[0024]

[0024] According to another aspect, the subject matter of the present disclosure deals with sensing the presence of worm-like features in a beam image.

[0025]

[0025] According to another aspect, the subject matter of this disclosure addresses detecting some types of optical misalignment from the presence of fringes parallel to and immediately adjacent to the edges of the near-field image.

[0026]

[0026] In accordance with another aspect, the subject matter of the present disclosure addresses the need to acquire images along the beam path in addition to near-field and far-field images.

[0027]

[0027] According to one aspect, a method for automatically detecting a condition of a light source for semiconductor photolithography is disclosed, the method comprising the steps of acquiring an image of a light beam generated by the light source, extracting a portion from the image, acquiring image data from the portion, filtering the image data using a matching filter, and storing results of the filtering operation in a database, the above steps being repeated over a period of time to repeatedly detect the condition.

[0028]

[0028] The method also includes comparing the plurality of stored results of the filtering operation to determine the evolution of a condition over time, and performing maintenance on the light source after a period having a duration based at least in part on the evolution of the condition over time. The condition can be the presence of a worm-like feature in the image. The condition can be the extent to which a slip surface defect exists on at least one optical element in the light source.

[0029]

[0029] According to another aspect, a method for automatically detecting the condition of a light source for semiconductor photolithography is disclosed, the method comprising the steps of obtaining an image of a light beam generated by the light source, extracting a portion from the image, obtaining image data from the portion, filtering the image data using a matching filter, appending a result of the filtering operation to a database containing a plurality of previous results of the above steps, comparing the plurality of results stored in the database to determine the evolution of the condition over time, and performing maintenance on the light source after a period having a duration based at least in part on the evolution of the condition over time.

[0030]

[0030] The condition can be the presence of a worm-like feature in the image. The condition can be the presence of a slip surface defect in at least one optical element in the light source. The condition can be the extent to which a slip surface defect is present in at least one optical element in the light source.

[0031]

[0031] The step of appending the result of the filtering operation to a database containing multiple previous results may be performed using wrap-around indexing. The step of appending the result of the filtering operation to a database containing multiple previous results may be performed using bit shifting. The method may include the additional step of generating video footage from the results stored in the database.

[0032]

[0032] According to another aspect, a photolithography apparatus is disclosed that includes a laser source for generating a laser beam and an imaging device for imaging the laser beam after it traverses an optical path of the light source, the optical path including a plurality of optical components, the imaging device including a screen element configured to receive at least a portion of the laser beam and generating an image of the laser beam when impinged by at least a portion of the laser beam, a camera positioned to view the screen element to obtain an image of the laser beam on the screen element, and an optical module positioned between the plurality of optical components and the screen element, the optical module including at least one movable optical element for altering the image on the screen.

[0033]

[0033] The optical module may include a variable focus lens. The optical module may include a parfocal lens. The apparatus may further include a driver, and the at least one moveable optical element may be mechanically coupled to the driver. The driver may include a stepper motor. The driver may include a piezoelectric transducer. The at least one moveable optical element may be mechanically coupled to a driver mounted on a track. The apparatus may further include a controller electrically connected to the driver, and the controller generates a control signal for controlling the driver. [Brief description of the drawings]

[0034] [Figure 1]

[0034] A schematic diagram, not to scale, of the overall broad concept of a photolithography system according to one aspect of the disclosed subject matter is shown. [Diagram 2]

[0035] 1 shows a schematic diagram, not to scale, of the overall broad concept of a lighting system according to one aspect of the disclosed subject matter. [Figure 3A]

[0036] 1 shows a schematic diagram, not to scale, of the overall broad concept of one example of a beam imager, Combined Automatic Shutter Measurement Module ("CASMM"). [Figure 3B]

[0037] 1 shows a schematic diagram, not to scale, of the overall broad concept of an example of a beam imager adapted to acquire images at multiple points along a beam path. [Figure 3C]

[0038] 1 shows a schematic diagram, not to scale, of the overall broad concept of another example of a beam imaging device adapted to acquire images at multiple points along a beam path. [Figure 4]

[0039] 1 shows a schematic diagram, not to scale, of a partial cutaway view of the overall broad concept of an electrically actuated through-wall adjuster according to one aspect of the disclosed subject matter. [Diagram 5]

[0040] 1 shows a schematic, partially cut-away view, not to scale, of the overall broad concept of a system for controlling a group of actuators of multiple electrically actuated through-wall adjusters, according to one aspect of the disclosed subject matter. [Figure 6]

[0041] 1 is a flowchart illustrating a method for aligning components in a lighting system according to one aspect of the disclosed subject matter. [Figure 7]

[0042] 1 is a pictorial representation of the overall broad concept of a GUI for controlling one or more actuators of a plurality of electrically actuated through-wall adjusters, according to one aspect of the disclosed subject matter. [Figure 8]

[0043] 1 is a flow chart illustrating a method for correcting aperture clipping according to one aspect of the disclosed subject matter. [Figure 9]

[0044] 1 is a flow chart illustrating a method for correcting center of gravity shift according to one aspect of the disclosed subject matter. [Figure 10]

[0045] 1 is a flow chart illustrating a method for correcting chamber resonances in accordance with one aspect of the disclosed subject matter. [Figure 11]

[0046] 1 is a pictorial representation of features that may be present in a CASMM near-field image. [Figure 12]

[0047] 1 is a flowchart illustrating a method for extracting features that may be present in a CASMM near-field image, according to one aspect of the disclosed subject matter. [Figure 13]

[0048] 1 is a flowchart illustrating a method for extracting features present in a CASMM image that may be indicative of a slip plane defect, according to another aspect of the disclosed subject matter. [Figure 14]

[0049] 1 is a flowchart illustrating a method for extracting worm-like features that may be present in a CASMM image that may indicate defects, according to another aspect of the disclosed subject matter. [Figure 15]

[0050] 1 is a flow chart illustrating a method of monitoring source optics for slip surface damage in accordance with another aspect of the disclosed subject matter. [Figure 16]

[0051] 10 is a flow chart illustrating a method of adding defect signatures to an accumulation array without storing the defect signatures individually, in accordance with another aspect of the disclosed subject matter. [Figure 17]

[0052] 10 is a flow chart illustrating a method for forming an image defect tracking array for generating video footage according to another aspect of the disclosed subject matter. [Figure 18]

[0053] 10 is a flowchart illustrating a method for updating an image defect tracking array using bit shifting, according to another aspect of the disclosed subject matter. [Figure 19]

[0054] 10 is a flowchart illustrating a method for updating an image defect tracking array using wraparound indexing in accordance with another aspect of the disclosed subject matter. [Figure 20]

[0055] 10 is a flowchart illustrating a method for detecting defects according to another aspect of the disclosed subject matter. [Figure 21]

[0056] 11 is a flow chart illustrating a method of viewing a generated beam image defect tracking array using wraparound indexing in accordance with another aspect of the disclosed subject matter. [Figure 22]

[0057] 6 is a flow chart illustrating a method for adding signals from an accumulation array according to another embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0035]

[0058] Various embodiments will now be described with reference to the drawings. In the drawings, like reference numerals are used to refer to like elements throughout. In the following description, for purposes of explanation, numerous specific details are set forth in order to facilitate a thorough understanding of one or more embodiments. However, it may be apparent that in some or all cases, the embodiments described below can be practiced without the specific design details described below. In other cases, well-known structures and devices are shown in block diagram form to facilitate the description of one or more embodiments. A simplified summary of one or more embodiments is presented below to provide a basic understanding of the embodiments. This summary is not an exhaustive overview of all contemplated embodiments, nor is it intended to identify key or critical elements of all embodiments or to delineate the scope of any or all embodiments.

[0036]

[0059] 1, there is shown a photolithography system 100 that includes an illumination system 105. As described more fully below, illumination system 105 includes a light source that generates a pulsed light beam 110 and directs the beam to a photolithography exposure tool or scanner 115 that patterns sub-microelectronic features on a wafer 120. Wafer 120 is placed on a wafer table 125 that is constructed to hold wafer 120 and connected to a positioner configured to precisely position wafer 120 according to certain parameters.

[0037]

[0060] The photolithography system 100 uses a light beam 110 having a wavelength in the deep ultraviolet (DUV) range, for example with a wavelength of 248 nanometers (nm) or 193 nm. The dimensions of the ultra-small electronic features patterned on the wafer 120 depend on the wavelength of the light beam 110, with shorter wavelengths resulting in smaller minimum feature sizes. For a wavelength of the light beam 110 of 248 nm or 193 nm, the minimum dimensions of the ultra-small electronic features may be, for example, 50 nm or less. The bandwidth of the light beam 110 may be the actual instantaneous bandwidth of the light spectrum (or radiation spectrum), which contains information on how the light energy of the light beam 110 is distributed across different wavelengths. The scanner 115 includes an optical arrangement, for example having one or more focusing lenses, a mask, and an objective arrangement. The mask is movable along one or more directions, for example along the optical axis of the light beam 110 or in a plane perpendicular to the optical axis. The objective arrangement includes a projection lens and enables image transfer from the mask to the photoresist on the wafer 120. The illumination system 105 adjusts the range of angles of the light beam 110 that strikes the mask. The illumination system 105 also homogenizes the intensity distribution of the light beam 110 across the mask.

[0038]

[0061] The scanner 115 may include, among other things, a lithography controller 130, an air conditioner, and power supplies for various electrical components. The lithography controller 130 controls how layers are printed on the wafer 120. The lithography controller 130 includes memory that stores information such as a process recipe. The process program or recipe determines the length of exposure on the wafer 120, the mask to be used, as well as other factors that affect the exposure. During lithography, multiple pulses of the light beam 110 irradiate the same area of ​​the wafer 120 to produce an exposure dose.

[0039]

[0062] Photolithography system 100 also preferably includes a control system 135. Generally, control system 135 includes one or more of digital electronic circuitry, computer hardware, firmware, and software. Control system 135 also includes memory, which may be read-only memory and / or random access memory. Suitable storage devices for specifically incorporating computer program instructions and data include, by way of example, all forms of non-volatile memory, including semiconductor memory devices such as EPROM, EEPROM, and flash memory devices, magnetic disks, such as internal hard disks and removable disks, magneto-optical disks, and CD-ROM disks.

[0040]

[0063] The control system 135 may also include one or more input devices (e.g., a keyboard, a touch screen, a microphone, a mouse, a handheld input device, etc.) and one or more output devices (e.g., a speaker or a monitor, etc.). The control system 135 also includes one or more programmable processors and one or more computer programs tangibly embodied in a machine-readable storage device for execution by the one or more programmable processors. Each of the one or more programmable processors can execute a program of instructions to perform a desired function by manipulating input data and generating appropriate output. Typically, a processor receives instructions and data from a memory. Any of the foregoing may be supplemented by, or incorporated in, specially designed ASICs (application-specific integrated circuits). The control system 135 may be centralized or partially or fully distributed throughout the photolithography system 100.

[0041]

[0064] With reference to Figure 2, exemplary illumination system 105 is a pulsed laser source that generates a pulsed laser beam as light beam 110. It will be apparent to one skilled in the art that Figure 2 depicts precisely one particular collection of components and optical paths for purposes of facilitating explanation of the broad principles of the present invention in general, and that the principles of the present invention may be advantageously applied to lasers having other components and configurations.

[0042]

[0065] 2 illustrates, in an exemplary block diagram, a gas discharge laser system according to an embodiment of certain aspects of the disclosed subject matter. The gas discharge laser system may include, for example, a solid-state or gas discharge seed laser system 140, an amplification stage, for example, a power ring amplifier ("PRA") stage 145, relay optics 150, and a laser system output subsystem 160. The seed system 140 may include, for example, a master oscillator ("MO") chamber 165 in which an electrical discharge between electrodes (not shown) induces a lasing gas discharge in a lasing gas to produce a population inversion of energetic molecules, including, for example, Ar, Kr, or Xe, to generate a relatively broadband radiation that may be line narrowed in a line narrowing module ("LNM") 170 to a relatively very narrow bandwidth and center wavelength selected, as known in the art.

[0043]

[0066] The seed laser system 140 may also include a master oscillator output coupler ("MO OC") 175, which may include a partially reflecting mirror that, together with a reflecting grating (not shown) in the LNM 170, forms an oscillator cavity in which the seed laser 140 oscillates to form the seed laser output pulses, i.e., forms the master oscillator ("MO"). The system may also include a line center analysis module ("LAM") 180. The LAM 180 may include an etalon spectrometer for fine wavelength measurements and a grating spectrometer for coarse resolution. The MO wavefront engineering box ("WEB") 185 may act to redirect the output of the MO seed laser system 140 to the amplification stage 145 and may include beam expansion, for example, using a multi-prism beam expander (not shown), and coherence destruction, for example, in the form of an optical delay path (not shown).

[0044]

[0067] The amplification stage 145 may include, for example, a lasing chamber 200, which may also be an oscillator formed, for example, by seed beam injection and output coupling optics (not shown), which may be incorporated into the PRA WEB 210 and may be redirected back through the gain medium in the chamber 200 by a beam reverser 220. The PRA WEB 210 may incorporate a partially reflective input / output coupler (not shown), a maximally reflective mirror for the nominal operating wavelength (e.g., about 193 nm for an ArF system), and one or more prisms.

[0045]

[0068] A bandwidth analysis module ("BAM") 230 at the output of the amplification stage 145 may receive the pulses of the laser light beam output from the amplification stage and take a portion of the light beam for metrology purposes, for example, to measure the output bandwidth and pulse energy. The pulses of the laser output light beam then pass through an optical pulse stretcher ("OPuS") 240 and an output coupling automatic shutter metrology module ("CASMM") 250, which may also be where the pulse energy meter is located. One purpose of the OPuS 240 may be, for example, to convert a single output laser pulse into a pulse train. Secondary pulses generated from the original single output pulse may be delayed relative to each other. By distributing the energy of the original laser pulse to the secondary pulse train, the effective pulse length of the laser may be expanded and at the same time the peak pulse intensity may be reduced. Thus, the OPuS 240 may receive the laser beam from the PRA WEB 210 via the BAM 230 and direct the output of the OPuS 240 to the CASMM 250.

[0046]

[0069] CASMM250 is a beam imaging / analysis tool that provides information about beam parameters. CASMM samples the laser beam to obtain both near-field and far-field profiles of the beam in a single camera image. The near-field image is the beam profile at the exit of the laser, while the far-field image provides information such as beam pointing and divergence. Distortions in the beam profile are captured by the near-field image and may be investigated to detect optical defects. This investigation requires advanced image processing techniques. Knowledge gained from CASMM images can be used to determine appropriate changes in the system, which can help maintain high beam quality.

[0047]

[0070] 3A is a highly simplified and generalized block diagram of a possible configuration of the CASMM 250. As shown, the CASMM 250 includes an automatic shutter 252 and additional instrumentation 254 used to monitor the laser performance in situ. The CASMM 250 receives an input beam that is split by a beam splitter 256. The automatic shutter 252 is positioned and configured such that when closed, it blocks the unsplit portion 258 of the laser beam and when open, it allows the laser beam to exit without interference. The automatic shutter 252 may include an uncoated CaF2 rotating prism using total internal reflection to direct the blocked beam to a water-cooled beam dump (not shown).

[0048]

[0071] The instrumentation 254 is configured to receive the split portion 260 of the beam and preferably includes various photodetectors and position detectors. As shown in FIG. 3B, the instrumentation 254 also preferably includes an optical system 264 and an image sensor 274, e.g., a 2D camera, that captures an image of the beam, which may be a 2D image of the near and far fields of the beam, but may also include intermediate images. Thus, one output of the CASMM 250 is a two-dimensional (2D) cross-section of the intensity of the beam profile. This 2D cross-section may be used to measure the beam profile and detect distortions or irregularities. The control system 135 collects data from the 2D camera in the instrumentation 254. The control system 135 processes the data to derive useful information for immediate observation of the beam polarization, profile, divergence, and pointing, as well as for long-term storage and retrieval. The immediate observation of the data may be part of a field service event (alignment, troubleshooting) of the laser. Storage and retrieval can be part of event investigation or trending of laser performance over time for preventive maintenance purposes.

[0049]

[0072] The image is acquired in a plane perpendicular to the beam propagation direction. A fluorescent glass screen 266 of appropriate illumination depth and decay time is used to down-convert 193 nm to visible wavelengths for camera acquisition to generate a visible image of the beam. The near-field and far-field 2D imaging is preferably simultaneous and continuous during operation of the laser. The near-field image can be an image of a plane containing the system hard aperture inside OPuS 240.

[0050]

[0073] The near field provides unique information about the irradiance profile, an important characteristic of the beam. Its distribution over the cross section is of great importance to the lithography industry. Slip surfaces, dendritic growth (deterioration of the beam splitter coating quality), misalignment, clamp-induced stresses, and dust particle contamination occurring in most beam splitters all have distinctive features that can be seen in the near field image or extracted by analysis of the near field image. Another image feature is a short line cluster caused by diffraction by a line defect. This is sometimes called "vertical line damage", but in reality the orientation of this feature depends on the orientation of the CASMM imager relative to a particular optical component in the PRA WEB and is not always vertical. This correlates with damage observed in the optical components in the form of relatively short lines oriented vertically, hence the name. Evidence of this damage appears in the CASMM images in the form of clusters of relatively short interference fringes.

[0051]

[0074] Beam imaging / analysis tools such as CASMM as previously implemented are useful for assessing parameters such as laser alignment and polarization ratio. They are also useful for assessing the health of optical components in the system such as optical pulse stretchers and beam splitters. However, such tools typically lack the ability to observe the beam at positions along the beam path prior to the optical pulse stretcher and beam splitter. As a result, during normal laser operation, if one or more optical components upstream of the optical pulse stretcher and beam splitter are suspected to be damaged, the location of the damage (i.e., module) cannot be pinpointed precisely because there is no way to observe the optical components or intensity distribution in the optical path in situ. Apart from damage in the optical pulse stretcher and beam splitter where an estimate of the amount of damage and preventive replacement of components is possible, there is no conventionally available method for tracking optical damage.

[0052]

[0075] To address these potential issues, it is beneficial to equip a beam imaging / analysis tool such as the CASMM250 with a lens that has a variable focal length. This can be, for example, a varifocal (variable focus) lens, which changes focus as the focal length (and magnification) is changed, or a parfocal ("true") zoom lens, which remains in focus as the lens zooms (changing focal length and magnification).

[0053]

[0076] In other words, in accordance with one aspect of the present invention, the instrument 254 of the CASMM 250 includes a system for dynamically adjusting the focal length of the lens before imaging the beam onto the phosphor screen 266. One such system is shown in Figure 3B. In Figure 3B, the optical system includes multiple focusing elements, including an afocal zoom system 264, which may be a combination of convex and concave prisms as shown. The components of the afocal zoom system 264 may be configured to move relative to a focusing lens that remains fixed.

[0054]

[0077] To effect movement of the components of the afocal zoom system 264, the lenses may be placed on a track 276, and the position of one or more lenses may be adjusted to change the focus. The position of the lenses may be moved using drivers 270a, 270b, and 270c, which may be, for example, steppers or piezoelectric transducer (PZT) drivers. Signals to the drivers 270a, 270b, and 270c may be provided by a local controller 272, which includes software that may interface with a user to receive commands for movement. The movement of the drivers 270a, 270b, and 270c may also be automated to routinely move optical elements and collect images to generate a history and characterize the behavior of the optical components upstream of the CASMM. By varying the focus, the images on the screen 266 acquired by the camera 274 will be from different positions along the optical path (e.g., OPuS 240, PRA WEB 210, etc.), allowing aspects such as the intensity distribution of light in these components to be analyzed. The local controller 272 may include presets that focus the afocal zoom system 264 at a preselected position along the beam path.

[0055]

[0078] The system, if equipped with a software or hardware crosshair target, can also be used to evaluate the position of the beam striking the optics at different locations along the laser path. Beams located near the edges of a prism can drift when the laser firing pattern is changed, potentially reducing energy and / or bandwidth efficiency.

[0056]

[0079] In another embodiment, one or more sets of fixed lens assemblies can be used in place of the afocal zoom system 264. This is shown in FIG. 3C, where instead of the afocal zoom system 264, there is a group 265 of fixed lens assemblies 265a, 265b, 265c, 265d, and 265e. Although the group 265 is shown as having five fixed lens assemblies, it will be apparent to one skilled in the art that the group 265 may have fewer or more fixed lens assemblies. Also, while each of the fixed lens assemblies is shown as one lens, it should be understood that each fixed lens assembly can be composed of several lenses, prisms, etc. The fixed lens assembly 265a can be a conventional fixed lens assembly that focuses on an output aperture, the fixed lens assembly 265b can be selected to focus on a first optical component that is upstream from the output aperture (i.e., earlier in the beam path), the fixed lens assembly 265c focuses on a second optical component that is upstream from the output aperture, and so on. A local controller 272 controls lens positioner 273 to position the desired fixed lens in the beam path. The movement of group 265 to selectively place one of the fixed lens assemblies in the optical path is preferably transverse to the optical axis (the direction of propagation of beam 260), in contrast to a variable focal length lens control system that is preferably configured to translate the optics along the optical axis.

[0057]

[0080] The above system improves the ability to predict optical failures, schedule service events to minimize interference with customer operations, and reduce unplanned laser downtime. As mentioned above, the system is preferably located in the instrumentation 254 of the CASMM 250. This provides the ability to focus at multiple desired positions in the beam path. In other words, the beam at this position has traversed substantially all of the optical components in the output subsystem, so that the instrumentation can substantially "see" many of these optical components from this position. The system may also be configured to focus at known positions of optical elements in the illumination system 105 through the use of pre-settings in the system shown in FIG. 3B or through the selection of fixed lenses in the system shown in FIG. 3C.

[0058]

[0081] Many of the modules described above include provisions for aligning the module and / or the optical components within the module. As used here and elsewhere in this specification, including the claims, "optical element" refers to a module or an optical component within the module. Similarly, incrementally adjusting an optical element means changing the position and / or orientation of the optical element in one or more negative or positive increments that are small compared to the overall adjustment range of the position or orientation of the optical element. As an example of an incrementally adjustable optical element, the LNM 170 may include provisions for adjusting the position of its vertical tip. The tip and / or tilt of the MO OC 175 is preferably adjustable. The MO WEB 185 may include a folding mirror with an adjustable tip or tilt. The BR 220 may include a prism with adjustable horizontal translation and tilt. The PRA WEB 210 may have an adjustable optical output coupler and / or a folding mirror with an adjustable tip or tilt.

[0059]

[0082] Typically, adjustments of these components are made using respective through-wall adjusters ("TWAs"), which require a manually operated hex driver to point, tilt, or translate the optic or module. The TWA may provide sealed mechanical feedthroughs for certain adjustments, which are accessed, for example, through a cover via the sealed mechanical feedthroughs.

[0060]

[0083] According to one aspect of the presently disclosed subject matter, adjustments are made with an electrically actuated TWA instead of a manually actuated TWA. A motor is mechanically coupled to the TWA. For example, the motor may have a shaft to which a hex adapter is attached, so that when the motor rotates the shaft, the hex driver also rotates, translating the end of the TWA along its axis according to the direction of rotation of the shaft. Preferably, the motor is a stepper motor that can be rotated in several discrete steps in either direction. The TWA preferably has a very fine (e.g., 1 / 4-80) thread pitch, so that the stepper motor requires a relatively large number of steps to move the TWA a short distance.

[0061]

[0084] Such a configuration is shown in FIG. 4, where the TWA 300 protrudes through a bushing 305 in the module wall 310 and is mechanically coupled to an optical feature 315 to control some aspect of the optical feature 315 (e.g., position, tilt, tip). The optical feature may be an optical element as described above, or all or part of an optical element. The TWA 300 is rotated about an axis by a motor 320 to cause the TWA 300 to extend or shorten. The motor 320 is mechanically coupled to the TWA 300 by a shaft 330. If the TWA 300 is of an existing design with a female hex socket, the shaft 330 may have a male hex driver tip. The motor 320 is secured to the module wall 310 by a fastening device 340, which may be secured to the bushing 305 as shown to limit the relative movement of the motor 320 and the TWA 305. Motor 320 receives control signals from control system 135 as shown, and is also connected to a power source (not shown).

[0062]

[0085] As shown in FIG. 5, the overall system may include several actuators 320. As mentioned above, the actuators 320 may be stepper motors, but other devices such as, for example, non-stepper motors or valves, pressure control devices, piezoelectric devices, linear motors, hydraulic actuators, voice coils, etc. may also be used. As shown, the actuators 320 are configured to be under the control of the control system 135. During alignment, for example, the control system 135 applies signals directly or indirectly to the actuators 320 to cause the actuators to place the respective optical features 315 in a properly aligned position and / or orientation. In some steps of the alignment procedure, the control system 135 causes the actuators to insert an optical element having a phosphor screen into the beamline and into the field of view of the miniature camera so that alignment to the element can be performed and verified. This alignment may be performed in any one of several modes of operation. For example, as described in more detail below, alignment may be performed in a manual mode, a semi-automatic mode, a fully automatic mode, a light source management mode, or a scanner management mode.

[0063]

[0086] For example, in manual mode, a user can adjust the individual actuators 320 via an input device 400 having a computer-based interface, such as a graphical user interface ("GUI"). In some embodiments, the GUI can be configured to display information about the beam profile as detected at one or more positions along the optical path in the laser source. These positions can be positions where near-field or far-field images are acquired, or additional positions can be used if the beam analysis and imaging tool is described elsewhere herein as having a variable focus. For example, the beam profile information profile can include a two-dimensional intensity map, other intensity information, beam width information, or other profile information. The beam profile information can be obtained by illuminating a phosphorescent surface with the beam or a portion of the beam and / or by receiving information from a metrology module (e.g., CASMM250, etc.). This information can be used, for example, to obtain diagnostic information and / or to manually adjust optical elements in the laser source. As another example, in semi-automatic mode, a combination of operations can be initiated with a single user input from the input device 400. Such operations may include, for example, an automatic sweep of a selected actuator to complete one of the tasks in an alignment sequence. As another example, in an automatic mode, an automatic performance of a series of tasks in an alignment protocol may be completed. In various implementations of a semi-automatic or automatic mode, one or more operations may be based at least in part on beam profile information. In various implementations of a semi-automatic or automatic mode, beam profile information may be collected and adjustments made based on a request signal received from a scanner receiving light from a light source.

[0064]

[0087] As another example, in light source management mode, one or more optical components may be adjusted based on image data from CASMM 250 in light source 105. For example, the adjustments may be based on correcting for diffraction fringes using CASMM 250 to monitor the beam. Here and elsewhere, adjustments are intended to include, among other things, realigning and / or repositioning lenses, mirrors, prisms, or other optical components, modifying the timing of an event or measurement, modifying the schedule of a maintenance event, or combinations thereof.

[0065]

[0088] For example, if the beam is diffracted from the aperture, fringes may be detected in the near-field image from the CASMM 250. These fringes can be suppressed, for example, by adjusting one or more mirrors using the electrically actuated TWA 320.

[0066]

[0089] A possible alignment procedure is shown in FIG. 6. In step S60, initial alignment position information is collected. In step S62, baseline data is collected. Various subsystems and optical components may then be aligned. For example, in step S64, the MO subsystem 140 may be aligned. In step S66, the folding mirrors in the MO WEB 185 and the PRA WEB 210 may then be aligned. In step S68, the prisms in the beam reverser 220 may then be adjusted. In step S70, an initial adjustment of the MOPRA may then be made. In step S72, an adjustment may be made using the folding mirrors to optimize the MO seed alignment. In step S74, the MOPA mode alignment may then be optimized. In step S76, the OPuS 240 may then be aligned. In step S78, the final alignment and specifications may then be verified. In step S80, the overall system performance may then be verified.

[0067]

[0090] As mentioned above, the alignment system may operate according to any one of several modes. In a manual mode, a user, such as a technician or field service engineer, may directly modify the position and / or orientation of the available optical components using an input device 410 in the form of a GUI as shown in FIG. 7. As shown, the GUI may include a graphical representation 420 of an electrically actuated TWA about to be adjusted, along with graphical representations 430, 432, and 434 of intensity profiles or other beam characteristics, such as images of phosphor screens, and along with a 2D image 440 from the CASMM 250. In the illustrated example, there are three graphical representations, although it should be understood that fewer or more graphical representations may be provided.

[0068]

[0091] The alignment system can also be operated in a semi-automatic mode, where the user can initiate a combination of operations with a single input. Such operations can range from including an automatic sweep of selected actuators to completing one of the tasks in the task sequence shown in FIG.

[0069]

[0092] It should be noted that it may be preferable to use not only the near-field image but also the far-field image for alignment purposes, which can be used to determine, for example, whether the light beam is properly oriented.

[0070]

[0093] FIG. 8 is a flow diagram of a process for correcting diffraction fringes from aperture clipping. Although the figure and the following description apply to a single optical component, it should be understood that the process can also be systematically and sequentially applied to two or more optical elements. In step S82, the current state of an optical element, such as a mirror, with respect to the available degrees of freedom (e.g., tip / tilt / translation) is recorded as a base setting. In step S84, a determination is made as to which edge of the near-field image shows diffraction fringes due to aperture clipping. Since the near-field image is rectangular, all four sides are checked for the presence of diffraction fringes. In step S86, it is determined whether diffraction fringes are present at two or more edges of the near-field image. If so, in step S88, it is determined whether fringes are present on opposite sides (vertical or horizontal) of the near-field image. If the determination in step S88 is yes, in step 90, it is assumed that simple adjustment of the tip or tilt of that optical element is insufficient to completely eliminate the aperture clipping because the beam is overfilling the aperture. Subject to the constraint that the energy centroid and pointing direction of the beam remain within specifications, small adjustments to the tip or tilt angles of the optics may be made to balance the extent of the diffraction fringes, after which the process ends. On the other hand, if step S88 determines that the diffraction fringes are not on opposite sides, but instead on adjacent edges, the process logic proceeds to step S92 with the knowledge that adjustments can be made in two directions. In step S92, if the evaluation of step S88 is that two adjacent edges show diffraction fringes, the image shift direction is horizontal and vertical, and if the evaluation of step S86 is that the diffraction fringes are only on one side, the image shift direction is horizontal or vertical. The process logic proceeds to step S94 where the optics are either pointed and tilted, or pointed or tilted according to the image shift direction derived in step S92. In step S96, it is determined whether the actions taken in step S94 have ameliorated the streaks, ie, suppressed the streaks (which indicates at least partial mitigation of aperture clipping).If yes, and step 98 determines that the improvement is sufficient, then the process ends. If step 98 determines that the improvement is not sufficient, then the process flow returns to step S94. If step S96 determines that the fringes have not been improved, then step 100 determines whether the process is expected to be successful. If yes, then the process flow returns to step S94. If not, then in step S102, the tip / tilt / translation states of the optical elements are returned to the base settings recorded in step S82, and the process ends. If the image shift direction determined in S92 indicates that two directions of adjustment are required (due to the presence of fringes on the adjacent edge determined in S88), then the operations described for S94, S96, S100, S102, and S98 are performed for each adjustment direction.

[0071]

[0094] As another example of possible operation in source management mode, adjustments may be based on correcting damage that induces a spatial bias in a beam parameter such as the beam centroid, e.g., the formation of a slip surface in an optical component. As in the previous example, a near-field image from the CASMM250 is analyzed. If there is an indication that a slip surface is forming, one or more optical elements may be adjusted to re-center the beam.

[0072]

[0095] FIG. 9 is a flow diagram of a process for compensating for the shift of the beam center of gravity due to a sliding surface. Although the figure and the following description apply to a single optical component, it should be understood that the process can also be systematically and sequentially applied to two or more optical elements. In step S202, the current state of an optical element, such as a mirror, with respect to the available degrees of freedom (e.g., tip / tilt / translation) is recorded as a base setting. In step S204, it is determined whether a sliding surface exists. This may be accomplished using a process such as that described in more detail below in connection with FIG. 12. In step S206, it is determined whether the sliding surface is at the center of the near-field image, or instead whether the sliding surface is offset from the center of the near-field image. If the sliding surface is at the center of the near-field image, step S208 determines to take no action and the process ends. If the sliding surface is not at the center of the near-field image, in step S210, it is determined whether the center of gravity of the near-field image is offset in the same direction as the sliding surface offset. If the center of gravity of the near-field image is not offset in the same direction as the offset of the slide surface, a decision is made in step S208 to take no action and the process ends. If the center of gravity of the near-field image is offset in the same direction as the offset of the slide surface, a shift direction of the slide surface is determined in step S212. The process logic proceeds to step S214, where the optic is pointed and tilted, or pointed or tilted according to the slide surface shift direction derived in step S212. In step S216, it is determined whether the action taken in step S212 improved the position of the center of gravity. If yes, and if step S218 determines that the improvement is sufficient, the process ends. If step S218 determines that the improvement is not sufficient, the process flow returns to step S214. If step S216 determines that the center of gravity has not improved, a decision is made in step S220 as to whether the process is expected to be successful. If yes, the process flow returns to step S214. If not, in step S222, the tip / tilt / translation states of the optics are returned to the base settings recorded in step S202, and the process ends.

[0073]

[0096] It should be noted that although this process is described as correcting for centroid shifts due to slip planes, the principles are not limited to correcting centroid shifts due to slip planes, and this process can also be advantageously applied to correct centroid shifts due to other factors, such as the appearance of white fluff on the beam footprint due to compression under high UV fluence on CaF2 optics.

[0074]

[0097] As another example of a possible operating mode, in a scanner-managed mode, the scanner can command the use of a particular repetition rate that is close to the chamber resonance frequency. In such a situation, the light source can respond by adjusting one or more optical components to improve pointing by reducing the pointing total inclusion range ("TIR"). FIG. 10 is a flow diagram of a process by which the light source compensates for over-pointing (i.e., the pointing TIR is greater than a specified value) when the scanner commands operation at a repetition rate that corresponds to the chamber resonance. This process includes pointing the tip of the PRA OC and / or tilting the PRA OC to compensate for the over-pointing. The resonance peak is lowered, the lower limit is raised, or a combination of the two is implemented to reduce the TIR. It is beneficial to use a look-up table to provide predetermined tip / tilt adjustment values ​​to minimize the time required to compensate for the effect of the resonance peak.

[0075]

[0098] More specifically, as shown in FIG. 10, in step S302, the current state of the optical elements, such as mirrors, for the available degrees of freedom (e.g., tip / tilt / translation) is recorded as a base setting. In step S304, it is determined whether the scanner has requested a new repetition rate (in the figure, "RR"). In step S306, it is determined whether the new repetition rate is near the resonant frequency of the chamber. If the new repetition rate is not near the resonant frequency of the chamber, in step S308, it is determined to take no action and the process ends. If the new repetition rate is near the resonant frequency of the chamber, in step S310, a pre-shift value is looked up in lookup table 312. This pre-shift value is selected to provide optimal compensation for the resonance. More specifically, the pre-shift value is a value determined a priori, i.e., through rigorous system characterization and production calibration. One way to do this includes performing a careful optical alignment and recording the positions of the adjustable optical components as a base setting. This process is performed offline, but is otherwise identical to the first step of Figure 10. Operating conditions such as repetition rate and duty cycle are varied over a range of use cases, and their effect on the performance parameters is recorded. Adjustments to selected optical components that improve the performance parameters, particularly near the resonant frequency, are then made, and the positions of each of the adjustable optical components are recorded. This process is repeated over a range of operating conditions to generate a look-up table.

[0076]

[0099] The logic of the process proceeds to step S314 where the optics are pointed and tilted, or are pointed or tilted according to the pre-shift value obtained in step S310. Step S316 determines whether the actions taken in step S314 have sufficiently improved the pointing. If yes, the process ends. If step S316 determines that the improvement is not sufficient, the process proceeds to step S318 where it determines whether the maximum number of allowed attempts to improve the pointing has been exceeded. If not, the process returns to step S314. If the maximum number of attempts has been exceeded, the process ends. As the module and optics age, the tip / tilt settings stored in the lookup table 312 may no longer provide optimal compensation for the resonance, so step S318 provides optimization of the compensation for the resonance effects. CASMM images may also be used to monitor the divergence and detune the OPuS 240 to improve divergence performance.

[0077]

[0100] Note that prior to this invention, adjusting the pointing required taking the tool out of production, which meant lost revenue for the chip manufacturer. One advantage of the present invention is that corrections can be made during downtime, for example between wafer exposures. Also, depending on which autoalignment mode is used, pointing corrections can be made during a tool down event, and in a much shorter time than would be required using purely manual adjustments.

[0078]

[0101] As described, a system according to one aspect of the present disclosure provides for automatic analysis, extraction, and processing of near-field CASMM images. This provides the ability to monitor the development of defects. Detection of very low contrast features such as lines and circular objects in images requires techniques beyond basic image processing. Laser beams are highly distorted, and the task of automatic detection of rings or lines is not straightforward. The problem is complicated mainly due to the fact that low contrast lines and rings lie on highly distorted beam profiles. However, tracking the intensity of low level features in CASMM images makes it possible to monitor the development of component failures, quantify image distortions associated with slip planes, and estimate different levels of component failure. It is also possible to correlate the quantified image distortions with CASMM log data and to identify new features associated with other degradation mechanisms.

[0079]

[0102] Some of the above processes used to determine in situ conditions, such as the presence of a slip surface, include the use of a combined automatic shutter metrology module 250, or CASMM 250. CASMM 250 is attached to the illumination source 105 at the exit port of the light source. Among its functions is the ability to acquire near and far fields in a single image. In the above example, the near field was used. By carefully inspecting the near field image, it is possible to perceive very faint straight lines at selected angles, for example ±15° or ±30°. These are shown diagrammatically in the depiction of the CASMM near field image 500 in FIG. 11 as feature 510. However, these lines are difficult to see and easy to miss. Also in FIG. 11, the presence of aperture clipping is expected to appear as a series of vertical or horizontal lines at one or more edges of the CASMM near field image 500. This is represented in FIG. 11 as feature 520. Also in FIG. 11, the presence of point-like defects would be expected to show up as a circular image or "ring" in the CASMM near-field image 500. This is represented in FIG. 11 as feature 530. As in the case of sliding surfaces, these features are difficult to see and easy to miss, and there is currently no way to automatically detect them, nor is it currently possible to monitor the progression of the defects that produce these features over time. FIG. 11 also shows the so-called "worm-like" features mentioned above, which are associated with optical damage to one or more components through which the laser beam passes. This is represented in FIG. 11 as feature 540.

[0080]

[0103] FIG. 12 shows a method that allows the slip surface line and ring to be detected separately. With reference to FIG. 12, the process on the left side of the flow chart is used to detect interference rings, and the process on the right side of the flow chart is used to detect slip surfaces. Step S120 in the first block of FIG. 12 is image acquisition, which includes taking an image from the CASMM 250 and extracting a near-field image. As described above, 2D images of the near and far fields are acquired in a plane perpendicular to the propagation direction of the beam. A fluorescent glass screen is used to generate a visible image of the beam. An image sensor, such as a 2D camera, acquires 2D images of the near and far fields of the beam. The near-field image is then processed in a pre-processing step S122 to detect interference rings. A similar pre-processing step S124 is performed to detect slip surfaces. The pre-processing step S122 for detecting interference rings is substantially the same as the pre-processing step S124 for detecting slip surfaces, except that the frequency response may differ depending on the type of feature to be detected. The image data is preferably first passed through an unsharp filter. An unsharp filter is a simple sharpening operator whose name comes from the fact that it enhances edges (and other high frequency components in an image) through the procedure of subtracting a smeared or smoothed version of the image from the original image. An unsharp filter is sometimes called the negative of a Laplacian filter. The image data is then passed through a bandpass filter that removes undesired frequencies. The frequency response of the bandpass filter may be empirically selected to obtain the clearest distortion in the output image. Thus, the frequency response that can be used to detect interference rings may differ from the frequency response used to detect linear features such as slip plane lines. The output of the bandpass filter has a histogram (pixel intensity value vs. frequency of pixel intensity value) that can be stretched to enhance the contrast of the image.

[0081]

[0104] Preferably, circular features in the near-field image are extracted using two techniques that use the Circle Hough Transform ("CHT") to extract and detect circular features. The CHT is a specialization of the Hough Transform. The goal of this technique is to find circles in an incomplete input image. The Circle Hough Transform relies on generating an image (accumulation array) based on the input image such that the value of each pixel corresponds to the likelihood that the center of a distortion ring is at that particular pixel. The generated image is processed to find a plausible location of the center and estimate the corresponding radius. The process of building the accumulation array is sometimes called voting. Thus, in FIG. 12, in step S126, a 2D accumulation array (or accumulator matrix) is built. Thus, the accumulation array is essentially an image with higher values ​​at pixels that are more likely to be the center of a distortion ring. The first step is to find the gradient field of the input image and use a threshold to find pixels that have gradient values ​​above this threshold. At this point, the accumulation array is initialized with zero values ​​at all pixels. Then, for each vector in the gradient field, we determine all pixels that are in the same direction of this vector and increment their values ​​in the accumulation array. The range of the pixel is determined by a fixed value related to the radius of the maximum expected ring radius. Pixels with high gradient values ​​correspond to strong edges in the input image.

[0082]

[0105] A different approach may be used to build the accumulation array. Instead of incrementing pixels within each vector in the gradient field, a template may be used. A template is essentially an image that contains the vectors in a fixed, predefined manner. To build the accumulation array, the center of the template is placed at a pixel (say X) in the gradient field. Then, dot products are taken between all the vectors and pixels in the template and the corresponding vectors in the gradient field, the values ​​are added, and the sum is stored as the value of pixel X in the accumulation array. If a ring were centered on pixel X with the same radius as the template, all the vectors would be aligned exactly and pixel X would have a very high value. It is preferable to use a single template for the entire image. This approach gives the flexibility to generate templates of different shapes or to use actual distortions from the laser in operation.

[0083]

[0106] In step S128, the maximum value is found, and in step S130, the radius of the circle is estimated.

[0084]

[0107] In another branch of the process, a multi-level accumulation array is constructed in step S132. In step S134, the local maximum is found for each level.

[0085]

[0108] In step S136, the results of the two CHT analyses are combined.

[0086]

[0109] For slip surface detection, the process flow again occurs in two branches. In one branch, a Standard Hough Transform ("SHT") is performed by first constructing an accumulation array in step S138. Local maxima are then identified in step S140. Line segments ends are then estimated in step S142. In another branch, a Radon transform is performed by first constructing a 2D Radon transform array in step S144. The Radon transform is a projection of image intensity along radial lines oriented at a particular angle. Each column of this transform represents a projection using a different angle. The result is then used to smooth, filter, and find local maxima in step S146. This information is used to estimate line segments ends in step S148. In step S150, the results of the two branches are combined. The results of steps S136 and S150 are then preferably combined in step S152 to generate a processed near-field image. The processed near-field image contains features of the extracted shape where the features are more clearly visible and can be more easily detected automatically.

[0087]

[0110] In the presence of multiple low-level features, the Radon transform (or the standard Hough transform) cannot provide robust detection results. In such cases, an alternative approach is to use a bank of matched filters to detect lines at specific angles. Each filter can be a finite impulse response ("FIR") filter to process the image.

[0088]

[0111] One advantage of the above process is that images can be retrieved and processed at selected intervals over a period of time. Over time, features 510 indicative of the presence of a skid surface are predicted to develop such that the formation of the skid surface becomes more widespread or pronounced. This information can be used to determine when a maintenance event should be scheduled.

[0089]

[0112] To obtain a quantified measure of the extent of slip-plane degradation, a binary counter array may be initialized to a preset value (e.g., zero) prior to the occurrence of a slip-plane. As new slip-planes appear in subsequent images, the signal from the new slip-plane may be added to the counter array. The counter array may be integrated on an image-by-image basis to form a one-dimensional, time-dependent, quantized slip-plane signal.

[0090]

[0113] Since each image is correlated with a measurable number of incident light pulses, this quantized slip surface signal can also be considered as a function of pulse count. This alternative signal provides information that allows the cumulative pulse-related degradation to be evaluated. It also allows the comparison of the optical degradation of low and high usage light sources.

[0091]

[0114] The binary nature of the counter array ensures that contributions from previously detected slip surfaces that are temporarily undetectable because they are not currently illuminated are not misidentified as "new" slip surfaces when they reappear. An added benefit of this processing is that it reduces the noise in the quantized slip surface signal, which facilitates its use to determine when to replace an optical component.

[0092]

[0115] The size of the counter can be equal to the number of pixels of the image size. This process results in a quantized slippage signal that increases over time until the optics with excessive slippage are replaced. Also, by adjusting the detection angle of the slippage detection method, some types of laser misalignment can be detected.

[0093]

[0116] Therefore, the processed near-field image can successfully distinguish slip planes from point defects and from diffraction leaving the aperture. Vertical line defects can also be detected using this method.

[0094]

[0117] Thus, a method for automatically detecting sliding surface defects may be envisioned that includes the steps of: (1) parsing and converting a file containing binary source image data into a suitable format for analysis, (2) locating the location of the near field (this location varies slightly from image to image and from CASMM to CASMM), (3) extracting the portion of the CASMM image that corresponds to the near field, (4) enhancing the near field image by sequentially applying an unsharp filter, a bandpass filter, and a contrast diffusion function to make the features more useful for automated analysis, (5) performing a Hough transform by building an accumulative array with values ​​higher at locations more likely to be along line edges, finding maxima, and estimating the ends of line segments, (6) using a Radon transform to first build, smooth, filter, and find minima in a Radon transform array, (7) combining the results of the Hough transform (step 5) and the Radon transform (step 6), and (8) communicating the combined results to a database (via diagnostic and / or configurable values, data logs, etc.).

[0095]

[0118] A method for automatically detecting point defects may include the following steps: (1) parsing a file containing binary source image data and converting it into a format suitable for analysis; (2) finding the location of the near field (this location varies slightly from image to image and from CASMM to CASMM); (3) extracting from the CASMM image a portion corresponding to the near field; (4) enhancing the near field image by sequentially applying an unsharp filter, a bandpass filter, and a contrast diffusion function to make the feature more useful for automated analysis; and (5) using a first modified circular Hough transform (CHT) to find values ​​at locations more likely to be along the ring. (5) constructing a multi-level cumulative array where values ​​are higher at locations more likely to be centers of interference rings, (6) using a second modified circular Hough transform to build, smooth, filter, and find local minima; (7) combining the results of the first modified circular Hough transform (step 5) and the second modified Hough transform (step 6) to build a cumulative array where values ​​are higher at locations more likely to be centers of interference rings; (8) finding local maxima and estimating the radius of the circles; and (9) communicating the results to a database (via diagnostic and / or configurable values, data logs, etc.).

[0096]

[0119] A method for automatically detecting slip planes and point defects may include the following steps: (1) parsing and converting a file containing binary source image data into a format suitable for analysis, (2) finding the location of the near field (this location varies slightly from image to image and from CASMM to CASMM), (3) extracting the portion of the CASMM image that corresponds to the near field, (4) enhancing the near field image to make the features more useful for automated analysis, (5) building a cumulative array with high values ​​at locations that are more likely to be centers of interference rings or lines, (6) finding local maxima and estimating circle radii and line lengths, (7) finding local maxima and estimating line segment ends, and (8) communicating the results (via diagnostic and / or configurable values, data logs, etc.) to a database.

[0097]

[0120] A method for retrieving and processing CASMM images for each laser source having a CASMM may include the following steps: (1) identifying a unique identifier for the source (e.g., a serial number (SN)), (2) identifying an acquisition date for the source, (3) retrieving a CASMM image file corresponding to the acquisition date from a memory source, e.g., from a file on a server, (4) parsing the file and extracting near-field image data, (5) processing the near-field image data to detect distortions (e.g., lines and circles), (6) summarizing the distortion information, (7) reporting the distortion information, and (8) repeating steps 1-7 for each source having a CASMM.

[0098]

[0121] A method for retrieving and processing a set of CASMM images for a light source having CASMM may include the following steps: (1) identifying a value (e.g., a serial number) unique to a particular light source; (2) identifying a first acquisition date for the light source; (3) identifying a last acquisition date for the light source; (4) retrieving selected CASMM image files from a memory source, e.g., from a file on a server, for a date range between and including the first acquisition date and the last acquisition date, each image having a timestamp; (5) for each image file retrieved in chronological order, analyzing the file and extracting near-field image data; (6) processing the near-field image data to detect distortions (e.g., lines and circles); (7) integrating distortion signals from the lines to form a quantized slip plane signal spanning the time from the first acquisition date to the timestamp of each image; (8) reporting the distortion information; and (9) repeating steps 1-8 for each light source having CASMM. After step (7), there may be a step of extrapolating the distortion information to estimate the time to next failure ("failover time").

[0099]

[0122] A method for retrieving and processing a set of CASMM images for a light source having CASMM may be construed as including the steps of: (1) identifying a value unique to a particular light source, (2) identifying a first acquisition date, (3) identifying a last acquisition date, (4) retrieving selected CASMM image files from a memory source, e.g., from a file on a server, for a date range between and including the first acquisition date and the last acquisition date, each image having a timestamp, (5) for each image file retrieved in chronological order, analyzing the file and extracting near-field image data, (6) processing the near-field image data to detect distortions (e.g., lines and circles), (7) collecting and summing the maximum individual line distortions for each near-field image to form a quantized slip plane signal spanning the time from the first acquisition date to the timestamp of each image, (8) reporting the distortion information, and (9) repeating steps 1-8 for each light source having CASMM.

[0100]

[0123] A method for generating a quantized slip surface signal may also be construed as including the steps of: (1) generating a sequence of fully processed CASMM images (in binary form, i.e., each pixel is either 0 or 1) that have been processed to reveal lines oriented consistent with the orientation of the slip surface, (2) initializing a binary counter array to zero prior to the occurrence of a slip surface, (3) adding signals from new slip surfaces to the binary counter array as they appear, (4) integrating the counter array image by image to form a one-dimensional, time-dependent, quantized slip surface signal, and (5) resetting the binary counter when substantially all of the slip surface indicators in the fully processed CASMM images have disappeared (and the light source remains in use).

[0101]

[0124] A method for generating a quantized slip surface signal may also be construed as including the steps of: (1) generating a sequence of fully processed CASMM images (in binary form, i.e., each pixel is either 0 or 1) that have been processed to reveal lines oriented consistent with the line damage orientation, (2) initializing a binary counter array to zero prior to the occurrence of a slip surface, (3) adding signals from new slip surfaces to the binary counter array as they appear, (4) integrating the counter array image by image to form a one-dimensional, time-dependent, quantized slip surface signal, and (5) resetting the binary counter when substantially all of the slip surface indicators in the fully processed CASMM images have disappeared (and the light source remains in use).

[0102]

[0125] The above-mentioned methods for detecting slip surfaces are sensitive to noise in the form of randomly oriented short lines that are misinterpreted as slip surfaces. It is preferable to distinguish these short lines from slip surfaces.

[0103]

[0126] Also, the above methods are not specifically directed at detecting the so-called worm-like features seen in some CASMM images. These features are associated with coating damage on the OPuS beam splitter that degrades optical performance. It would be desirable to have advance information that the OPuS beam splitter is beginning to fail, so that replacement of the OPuS beam splitter can be scheduled to occur at the same time as other maintenance to minimize unplanned downtime.

[0104]

[0127] Also, in some methods, a large number of CASMM images must be processed each time one wishes to generate an optical image defect signal representative of the progression of a particular defect as a function of time. It is preferable to reduce the number of images that must be processed in each case, thereby reducing the computation time required to generate an updated optical image defect signal. This is achieved by using already processed images and performing an incremental update. In practice, data is preprocessed up to a time T1, for example last week, and the results of this preprocessing are stored. Then, if one wishes to update these preprocessed results with additional (for example this week) data, it is necessary to process only the additional data and then take the previous processing results and add the processed additional data. This incremental update is much faster than processing all the data (both previous and additional) in their entirety again.

[0105]

[0128] Another problem with using beam image analysis tools is that images must be accumulated over a period of time. Changes in beam direction must be taken into account to obtain a defect signal. Detection results of defect features such as slip surfaces and / or vermiform features must be accumulated over a period of time to account for the effects of changes in beam pointing from one image (day) to the next. Changes in beam direction lead to non-uniform illumination that affects the detection results. Changes in beam pointing are caused by several factors including repetition rate and duty cycle. Accumulation of detected features evens out the non-uniformity caused by changes in beam pointing. The accumulation results in a monotonically increasing function, which also helps to estimate the size of vermiform features in CASMM images. However, there is a variable offset between images, which makes the correction for changes in beam direction unreliable.

[0106]

[0129] It is important to calculate the integrated signal effectively, as this plays a key role in determining whether or not we will get a noisy signal. In previous methods, each image studied was different in size, due to the nature of the images acquired by CASMM. Therefore, to accumulate the signal over a period of time, all images were enlarged or reduced to match each other. This can result in an irregular integrated signal. This problem can be solved by ensuring that all CASMM images studied are of the same size. To do this, we define the analysis region by selecting a center pixel pxlctr and also an array whose x coordinate ranges from [pxlctr-nx:pxlctr+nx] and whose y coordinate ranges from [pxlctr-ny:pxlctr+ny]. To make the resulting array square, we choose nx and ny to be the same value.

[0107]

[0130] In other words, the black and white images obtained by applying the defect detection methods disclosed herein are used to quantify the extent of damage to the optical element under consideration. As an example, the number of pixels that are white in color (white pixels indicating the presence of a defect feature) may be used to estimate the extent of damage. In other words, a light source with more optical damage will result in a CASMM image with a larger number of white pixels compared to a CASMM image obtained from a light source with less optical damage.

[0108]

[0131] In general, it is desirable to analyze several CASMM images. Analysis of a single CASMM image is generally not sufficient, since it is necessary to take into account the effects of beam bouncing. Therefore, once a defect is detected, it is preferable to analyze a series of images. This can be achieved by accumulating the respective defect features from each of the images in the series. Accumulating defect features from several CASMM images in this way requires that the size of the CASMM images is the same, in the sense that the number of rows and columns must be the same. However, the near-field images extracted from CASMM do not always have the same size due to illumination variations.

[0109]

[0132] One method of making all CASMM images the same size involves using a magnification value to enlarge or shrink the images. However, this method has the drawback that it also enlarges or shrinks the defect features, thus resulting in an erroneous estimation of the accumulated defect signal. In general, this method results in an inflated value of the defect signal.

[0110]

[0133] Another method to make all CASMM images the same size involves finding the centroid of the image and taking a preset number of columns to the left and right of the centroid and a preset number of rows above and below the centroid for every CASMM image. Note that the preset values ​​can be changed depending on the light source if necessary. This method ensures that all images are the same size without shrinking or enlarging the image and the defect features contained in it. CASMM images that are too small in the sense of lacking sufficient rows and columns may be discarded as corrupted CASMM images. This method allows a more reliable quantification of the extent of optical damage in the light source.

[0111]

[0134] In some configurations described above, the progression of the defect is determined as a function of elapsed or operating time, however, there may be situations where it is desirable to determine the progression of the defect as a function of other parameters, such as frame pulse count, module pulse count, etc.

[0112]

[0135] Additionally, one technique for generating an updated optical image defect signal representative of the progression of a particular defect involves adding new image data as it becomes available and analyzing the entire data set, including images that have already been analyzed at least once, which can be a very time consuming process.

[0113]

[0136] It is possible to analyze a set of beam images and generate a separate defect signal for each type of optical defect for which a detection method exists. Updating the entire set would take even more time.

[0114]

[0137] Thus, there is a need for methods to detect and identify low level features and to attribute these features to specific defect types. A further need exists for methods to monitor and efficiently update the evolution of these features that can predict how much time remains before the optical performance of the light source degrades to the point that an optical component or module must be replaced.

[0115]

[0138] FIG. 13 shows another slip surface detection algorithm that is particularly insensitive to noise, for example noise resulting from the presence of short lines. A typical CASMM image has a small bright spot, which is the far-field image, and a rectangular field, which is the near-field image. A CASMM image is read in step S160, similar to step S120 in FIG. 12. In step S162 to extract the near-field of the CASMM image, the entire image is converted to a binary format. This conversion is performed by comparing each pixel to a threshold intensity value. All pixels with an intensity above this threshold are assigned a value of 1 (white), and other pixels are assigned a value of 0 (black). Each pair of consecutive white pixels is labeled. The pair with the maximum number of white pixels defines the near-field range. Extracting the locations of these pixels gives the near-field image. This scheme helps to detect near-field components in low-quality CASMM images. In step S164, an unsharp filter is applied, as described above in connection with step S124 in FIG. 12. In step S166, a passband filter is applied, also as described above in relation to step S124 of Figure 12. In step S168, the gradient of the filtered image is obtained.

[0116]

[0139] In step S170, the gradient of the filtered image is matched filtered against the gradient of a template. The template is defined in step S172, and the gradient of the defined template is obtained in step S174. As mentioned above, the template can be, for example, a series of lines that are longer than a predetermined length and at an angle of 30 degrees or more from vertical.

[0117]

[0140] The matched filtering applied in step S170 is used in signal processing to improve the signal-to-noise ratio (SNR). It is a process of correlating a template with the unknown signal. The template is a known signal that is expected in the observed signal, i.e., embedded in the CASMM image and not easily observed. This allows to detect the presence of template-like parts of the CASMM image. In other words, the template can be a generic template of the expected signal. It is correlated with the observed signal to improve the SNR. The occurrence of a peak in the matched filtered signal indicates the presence of the template in the signal. The location of the peak indicates the location of the template in the signal.

[0118]

[0141] Thus, in the presently described embodiment, slip surface detection is performed by matching the filtered CASMM image with templates. This process produces an image showing the correlation of each pixel in the CASMM image with the template. If a slip surface feature is present in the image, the correlation value at the corresponding location will be high. In step S176, a correlation threshold is applied. A threshold that gives unambiguous results can be empirically determined.

[0119]

[0142] Noise is classified as lines shorter than a certain line length threshold. This threshold can be determined by examining the slip planes and their lengths. For example, evaluation of the slip plane lengths from many different CASMM images suggests that to define a line that qualifies as a slip plane, the line in the image must be longer than 10 consecutive pixels. Thus, the detection process may exclude lines shorter than 10 pixels in length. This culling process is performed in step S178. The results are stored in an accumulation matrix in step S180. In step S182, the method determines whether all templates have been checked. If not, the template orientation is changed in step S184, and this new template orientation is then used for matched filtering. The process of FIG. 13 ends when all templates have been checked.

[0120]

[0143] Figure 14 shows a method for detecting worm-like features. As mentioned above, worm-like features appear in CASMM images and can be attributed to defects in upstream optical components, such as a defective OPuS beam splitter. Matched filtering is performed as described above in relation to Figure 13.

[0121]

[0144] The matched filtering process produces an output whose values ​​have been normalized based on the highest and lowest numerical values ​​in the result. This normalization process causes each pixel to have a value ranging between 0 and 1 inclusive. In a grayscale colormap, 0 is represented as black and 1 is represented as white. Values ​​between 0 and 1 are mapped to intermediate grayscale colors having values ​​between 0 and 1. For graphical representation, these intensity values ​​are mapped to a colormap. A convenient colormap is grayscale. Thus, if the intensity value is 0, the pixel is represented as black. If the intensity value is 1, the pixel is represented as white. Values ​​between 0 and 1 are mapped to intermediate grayscale colors.

[0122]

[0145] Due to the presence of noise, the detection of the worm-like features is preferably performed using a variable threshold. Initially, an initial threshold is kept high to confirm the presence of the worm-like feature. Once the detection of the worm-like feature exceeds this value, the presence of the worm-like feature is considered confirmed and a trace of the worm-like feature is added to the accumulation array. After the first detection and recording of the worm-like feature, the threshold is lowered. This allows absorbing effects due to beam bounce, i.e. the beam may not show the worm-like feature because it is not sufficiently irradiating the spot of the optical component that has a defect that causes the worm-like feature. Once the worm-like feature disappears (after a maintenance event), the detection algorithm will no longer detect any worm-like features and the value returned by the algorithm will be much smaller than the lower threshold used. A completely black image is recorded in the accumulation array. Once this happens, the threshold is returned to the higher value that was initially used.

[0123]

[0146] In FIG. 14, in step S172, the vertical line template for worm-like feature detection is rotated approximately 10 degrees to either side of the vertical to capture the entire feature. In step S176, it is assumed that pixels with intensity values ​​above 0.8 indicate the presence of a worm-like feature. In step S186, small speckles are rejected as noise. The process of FIG. 14 also includes an optional step S188 to filter out circles using template matching. In some cases, this is necessary to provide an error-free detection. Often, this is not necessary because circles are rare. The process of FIG. 14 also includes step S190 to fill in the areas near the left and right edges of the image with black. This step is done to ensure that any diffraction fringes that may be present are not interpreted as worm-like features.

[0124]

[0147] FIG. 15 shows an example of a method for monitoring the optics of a light source for slip surface damage using the slip surface detection method described above. Step S400 obtains a list of lasers from database 403 and defines a laser counter to uniquely associate a numerical sequence with each laser in the list of lasers. Step S402 determines whether all lasers on the list have been processed. If not, step S404 increments the laser counter to process the next laser. Step S406 retrieves the history accumulation array for that laser from memory 405. If no history accumulation array exists, the accumulation array is initialized. Step S408 applies the slip surface detection method to the earliest image not included in the accumulation array. Step S410 determines whether the detection method indicates an excessive number of slip surface defects. If the determination result of step S410 is yes, step S412 appends the slip surface signal to a history slip surface accumulation array in memory 407, which may or may not be separate from memory 405. If the answer at step S410 is no, the process returns to step S402. Step S414 determines whether there are additional images to process. If the answer is yes, the process returns to step S408. If the answer is no, then in step S416 the laser is flagged as having at least one, and possibly more than one, optical element with sliding surface damage, and an indicia of the flagged laser is stored in database 403. This process continues until all lasers on the list have been examined. Step S418 generates a report summarizing the defects found for all lasers on the list.

[0125]

[0148] The process of the method illustrated in Figure 15 is intended to analyze one defect type at a time. Although the specific example of a defect referenced in Figure 15 is a sliding surface, it will be apparent that the process of Figure 15 can be adapted to address other types of defects, such as worm-like features.

[0126]

[0149] Figure 16 shows a process for adding CASMM defect features to an accumulation array without storing the defect features individually. The process of Figure 16 handles multiple defect types that may include slip surfaces, worm-like features, interference fringes due to aperture clipping, interference fringes due to line damage, etc. The process of Figure 16 can be extended to as many types of features as one wishes to analyze and store.

[0127]

[0150] In FIG. 16, steps that are performed substantially the same as those in FIG. 15 are similarly numbered. In FIG. 16, step S420 retrieves an image from the current laser from database 403. This image can be the most recent image from the laser or another image. In step S407, a historical accumulation array for each member of the set of M types of defect features is retrieved from memory 409. In step S422, a defect detection method is performed to determine whether a defect feature is present in the image for each member of the set of M types of defect features. In step S424, the defect feature signal from the current image is added to the accumulation array. Optionally, data may be removed from the array to limit the time range under consideration to a predefined window defined in terms of time and / or number of pulses, e.g., 10 weeks, 20 weeks, etc. In step S410, it is determined whether the defect feature signal for each member of the set of M types of defect features exceeds a respective threshold value. If it is desired to monitor the source optics for multiple damage mechanisms, step S416 can be replaced or supplemented by generating and optionally storing a report, including setting flags in a database to indicate which defects were found.

[0128]

[0151] In one possible embodiment, an accumulation array may be implemented, with a binary accumulation array for each defect type, with each array address (pixel) storing either a 0 or a 1. If more granularity is desired in distinguishing between levels of intensity, the accumulation array may be configured such that each array address stores more than one bit, i.e., each address has a pixel depth, i.e., e.g., 8 bits, 16 bits, etc. In this manner, for example, 8-bit pixels may be used to assign grayscale levels as described above. Furthermore, multiple-bit pixels may be used to encode binary data for more than one type of feature. For example, an 8-bit pixel may be used to store binary data for eight different types of features, e.g., bit 0 may be used to indicate sliding surface damage, bit 1 may be used for worm-like features, etc. Also, multi-word pixels may be used. For example, a 32-bit pixel may be taken as 4×8 bits, which may be used as just described. For example, an 8-bit pixel could be used to store 8 bits of data for four different types of features, e.g., bits 0-7 could be used to indicate the level of slideway damage, bits 8-15 could be used to indicate the level of worm-like feature presence, etc. It will be apparent that the memory space could be configured to store a combination of binary (1-bit) representations for some features and word (multiple-bit) representations for another feature or features.

[0129]

[0152] FIG. 17 shows a method for forming an image defect tracking array that generates a video image. In step S224, a lookback window is defined. In step S226, an image counter is initialized, n=1 corresponding to the first interval of the lookback window (e.g., one day), and n=1, 2, ..., N, where N is the number of CASMM images that will be sampled within the lookback window. In step S228, a defect accumulation array DefectTypeMArray is initialized for each type of defect m under consideration, m=1, 2, ..., M, where M is the total number of detectable defect types, and each array is Nx x Ny x N, where Nx and Ny are the number of pixels in the near field of the CASMM image in the x and y directions. In step S230, an image n is taken and the respective counter and accumulation array for each defect type are initialized. In step S232, the data from sample n is evaluated for the presence of the defect under consideration, and all bits in plane n of DefectTypeMArray are set to 1 for all pixels where defect m exceeds a threshold. In step S234, the method determines whether the number of pixels in the data for sample n exceeds a predetermined threshold for the particular defect under consideration. If the answer to the question in step S234 is yes, then in step S236, a flag is set indicating the presence of that defect in the image corresponding to the data. Thereafter, or if the answer to the question in step S234 is no, then in step S238, it is determined whether all types of defects have been evaluated. If the answer to the question in step S238 is no, then in step S240, a defect type counter is incremented and the process returns to step S232. If the answer to the question in step S238 is yes, then in step S242, an image counter is incremented. In step S244, the method determines whether all images have been analyzed. If the answer to the question in step S244 is no, then the process returns to step S230. If the answer to the question in step S244 is yes, then in step S246, the defect type array is converted to a video with the image counter n as the time index. If desired, the time index can be replaced with a pulse count index.Note that the accumulation matrix is ​​three-dimensional, and each image corresponds to a particular two-dimensional array within the three-dimensional accumulation matrix.

[0130]

[0153] FIG. 18 illustrates how to update an image defect tracking array using bit shifting. Step S250 identifies the laser SN for which we wish to update the CASMM image defect tracking array. Step S252 initializes the defect type counter. Step S254 retrieves the defect type array DefectTypeMArray for defect type m. This array has a size of Nx×Ny×N. Step S256 retrieves the latest CASMM image of the laser ("New"). Step S258 analyzes the CASMM image for the particular type m of defect under consideration. Step S260 shifts the values ​​of bits in plane n of DefectTypeMArray for all Nx×Ny×N bits to plane n-1, for n=2:N, thereby erasing bits in plane 1 and making room for new samples in plane N. Since DefectTypeMArray is a 3D matrix, there is no carryover of results. In step S262, for the new data, the image is evaluated for the presence of defect m, and for all pixels where the defect type signal exceeds a predefined threshold, the bit in the plane is set to 1. In other words, for the data from "New N", the image is evaluated for defect m, and for all pixels where defect m exceeds the threshold, the bit in plane N of the DefectTypeMArray is set to 1. In step S264, all planes where the threshold is exceeded are flagged. The remaining steps of the method illustrated in FIG. 18 are the same as those described above for the like-numbered steps.

[0131]

[0154] FIG. 19 illustrates a method for updating an image defect tracking array using the wraparound indexing method described below. The method steps illustrated in FIG. 19 are the same as those described above for steps with the same reference numbers. Additionally, in step S270, image data is analyzed and stored according to the method described below. In step S272, it is determined whether a maximum of n has been reached, i.e., all images have been analyzed and their data stored. If the determination in step S272 is no, n is incremented by 1 and the process returns to step S270. If the determination in step S272 is yes, a determination is made as to whether all defect types have been evaluated as outlined above.

[0132]

[0155] One method of implementing step S270 as described above is shown in FIG. 20. As shown, step S280 determines whether a defect type array exists for the particular defect type under consideration. If the answer is no, step S282 creates an array. If the answer is yes in step S280, step S284 retrieves the defect type array. In step S286, a pointer is established or retrieved. The array pointer points to which plane in the DefectTypeMArray is filled, i.e., the next available plane. As will be described below in conjunction with FIG. 21, the array pointer points to which plane in the array is filled. In other words, the array pointer points to the next available plane. In step S288, a wraparound indicator is retrieved or established. In step S290, the most recent image of the laser under consideration is retrieved. In step S292, this image is analyzed for the defect type under consideration. Step S294 evaluates defect m for the data from "new N" and sets a bit in plane "new N" of DefectTypeMArray (pointed to by the array pointer) to 1 for all pixels in plane "new N" where defect m exceeds a threshold. Step S296 stores information related to this plane. This information may include laser identification, date the image was acquired, array pointer, wrap-around indicator, frame pulse count, etc. Step S297 appends the result of the store operation of step S296 to an accumulation array. In the illustrated example, the size of the accumulation array is determined by the size of the lookback window as described above. Step S299 determines whether the array pointer indicates a plane identification greater than the number of lookback window samples. If the answer to the question in step S299 is yes, step S298 sets the array pointer to 1 (or zero if zero-based indexing is used), so that the data in plane 1 is overwritten, and the process then ends. Step S300 then sets the wrap-around indicator. If the answer to the question in step S299 is no, the process ends.

[0133]

[0156] FIG. 21 illustrates a method according to another embodiment of the present invention for viewing CASMM image defect tracking arrays generated using wraparound indexing. In step S320, a laser for which one wishes to view the CASMM image defect tracking array is identified. In step S322, the defect type array for that laser is retrieved. In step S324, information associated with the defect type array is retrieved as well. In other words, information associated with the DefectTypeMArray is retrieved, such as laser SN, CASMM image date, array pointer, wraparound indicator, etc. In step S326, it is determined whether the wraparound indicator is set. If the determination in step S326 is yes, in step S328, a set of cumulative arrays is extracted for each plane from the plane indicated by the array pointer plus 1 to the end of the array. This generates a first set of planes. In step S330, a set of cumulative arrays is extracted for each plane from plane 1 to the plane indicated by the array pointer. This generates a second set of planes. In step S332, the first set of planes and the second set of planes are combined to form a third set of planes that are contiguous in time or in the order in which the samples were taken. If the determination in step S326 is no, in step S334, an alternate cumulative array is extracted for each plane, from 1 to the plane indicated by the array pointer. After step S334 or step S332, step S336 is performed, in which the defect data is presented in a desired format, e.g., a movie, a line plot, a comparison with laser parameters such as polarization, pointing, emittance, module pulse count, etc. The process then ends.

[0134]

[0157] FIG. 22 illustrates a method according to another embodiment of the present invention for adding signals from an accumulation array. In step S340, the accumulation array for a given laser is taken and a defect index m for the defect under consideration is initialized. In step S342, an accumulated defect signal DefectM is initialized with the first array of the accumulation arrays for the given defect type under consideration. In step S344, the laser index n is incremented and an array for that laser is taken from the accumulation array. In step S346, it is determined whether the magnitude of the defect signal is less than a predetermined threshold. If the answer to the question in step S346 is yes, in step S348, the accumulated defect signal is reset to zero. If the answer to the question in step S346 is no, in step S350, the taken signal is added to the accumulated defect signal. In either case, in step S352, it is determined whether the index for that laser is equal to the size of the accumulation array. If the answer to the question in step S352 is no, the process returns to step S344. If the determination in step S352 is yes, then in step S354 a determination is made as to whether processing has been completed for all defect types of interest, m=1,...,M. If the determination in step S354 is no, then in step S356 the defect type index m is incremented and the process returns to step S342. If the determination in step S354 is yes, then the process ends.

[0135]

[0158] As mentioned above, in the process described in connection with Figure 21, in step S326 a determination is made as to whether the wrap-around indicator is set. The wrap-around indicator simply indicates whether the available N planes of data corresponding to the lookback window are full. The diagram below shows the four states for a linear array of size N. [Table 1]

[0136]

[0159] In the leftmost array, new data is added to an empty location (plane) in the array. This form of additive indexing is not wraparound indexing, so the wraparound indicator is not set to true. Instead, the stored wraparound indicator has a value of false indicating that wraparound indexing is not being used. Determining whether the wraparound indicator is set is determining whether the indicator is set to true. In another example of this diagram, new data is added to an already occupied location in the array. Because this is wraparound indexing, the wraparound indicator is set to true.

[0137]

[0160] As shown, values ​​other than a simple sequential indicator can be used as the index n, for example a timestamp can be used as the index, as shown below: [Table 2]

[0138]

[0161] Alternatively, the frame pulse count can be used as the index, as shown below. [Table 3]

[0139]

[0162] Any combination of these values ​​can also be used as an index. The index can also be used to track additional related information. For example, if a timestamp is used as the index, then a laser number, frame pulse count, and other information can be associated with that timestamp, as shown below. [Table 4]

[0140]

[0163] As mentioned above, one aspect of the invention is a method that can reduce noise when detecting slip surface features in a source near-field image. Another aspect is a method that can detect worm-like features from a near-field image of a source optic. Another aspect is a method that can detect features of multiple degradation mechanisms in a source near-field image. Another aspect is making the defect features more reliable to help compare the signal to laser performance, resulting in a method to quantify laser damage based on the defect signal. Another aspect is a method to store the defect features in an array, which is computationally more efficient than generating a video every time.

[0141]

[0164] The above description is primarily in terms of analyzing near-field images of a beam analysis tool such as CASMM for illustrative purposes only. It will be apparent to those skilled in the art that the above method can also be applied to find features in beam images acquired at locations along the beam path other than those used to acquire the near-field image.

[0142]

[0165] Some examples of what making an adjustment may mean include, among others, realigning and / or repositioning lenses, mirrors, prisms, or other optical components, modifying the timing of an event or measurement, modifying the schedule of a maintenance event, or combinations thereof.

[0143]

[0166] The above description includes examples of several embodiments. Of course, for purposes of describing the foregoing embodiments, it is not possible to describe every conceivable combination of components or methods, but one skilled in the art can appreciate that numerous further combinations and permutations of the various embodiments are possible. Accordingly, the described embodiments are intended to embrace all such changes, modifications, and variations that fall within the spirit and scope of the appended claims. Furthermore, to the extent that the term "includes" is used in either the detailed description or the claims, such term is intended to be inclusive in a manner similar to the term "comprising," as "comprising" is interpreted when used as a transitional term in the claims. Furthermore, although elements of the described aspects and / or embodiments may be described or claimed in the singular, the plural is also contemplated, unless limitation to the singular is expressly stated. Furthermore, unless otherwise specified, all or a portion of any aspect and / or embodiment may be utilized with all or a portion of any other aspect and / or embodiment.

Claims

1. a light source having at least one optical element positioned within a housing; an actuator extending through a wall of the housing and mechanically coupled to the at least one optical element, the actuator adapted to change a position and / or orientation of the at least one optical element in one or more small negative or positive increments that compensate for optical degradation relative to an overall adjustment range for alignment of a position or orientation of the at least one optical element in response to a signal, the actuator having components for imparting motion to the actuator in response to the signal located outside the housing; a module arranged to acquire a near-field image of a light beam from the light source, the module having an instrument arranged to extract information from the near-field image; a control system for generating the signal at least partially in response to the information extracted from the near-field image; the at least one optical element comprises at least one of a master oscillator output coupler, a master oscillator wavefront engineering box folding mirror, a beam reverser prism, a power ring amplifier wavefront engineering box output coupler, and a power ring amplifier wavefront engineering box folding mirror; the actuator is adapted to at least one of: a) changing the tilt of the output coupler of the master oscillator; b) adjusting the tip or tilt of the fold mirror of the master oscillator wavefront engineering box; c) adjusting the horizontal translation and tilt of the prism of the beam reverser; d) changing the tilt of the output coupler of the powering ring amplifier wavefront engineering box; and e) adjusting the tip or tilt of the fold mirror of the powering ring amplifier wavefront engineering box. Photolithography equipment.

2. The photolithographic apparatus of claim 1 , wherein the control system generates the signals at least in part in response to commands entered by a user.

3. the modules further configured to obtain a far-field image of the light beam from the light source, and further including a module configured to extract information from the far-field image; 2. The photolithographic apparatus of claim 1, wherein the control system generates the signal at least in part in response to the information extracted from at least one of the near field image and the far field image.

4. 2. The photolithographic apparatus of claim 1, wherein the module is arranged to extract information from the image, the extraction of information being indicative of a state of alignment of the optical elements.

5. the control system generating the signal based on the information indicative of a state of alignment of the optical element; The photolithographic apparatus of claim 4 , wherein the actuator, mechanically coupled to the optical element, moves the optical element in response to the signal.

6. a light source having at least one optical element; a module arranged to acquire near-field and far-field images of a light beam from the light source, the module having a metrology instrument arranged to extract near-field information indicative of an alignment of the optical element from the near-field image and to extract far-field information indicative of an alignment of the optical element from the far-field image; a control system for generating a control signal based at least in part on at least one of the near field information and the far field information; an actuator mechanically coupled to the at least one optical element and responsively connected to the control system, the actuator adapted to change the position and / or orientation of the at least one optical element in one or more small negative or positive increments to compensate for optical degradation compared to an overall adjustment range for alignment of a position or orientation of the at least one optical element in response to the control signal; the at least one optical element comprises at least one of a master oscillator output coupler, a master oscillator wavefront engineering box folding mirror, a beam reverser prism, a power ring amplifier wavefront engineering box output coupler, and a power ring amplifier wavefront engineering box folding mirror; the actuator is adapted to at least one of: a) changing the tilt of the output coupler of the master oscillator; b) adjusting the tip or tilt of the fold mirror of the master oscillator wavefront engineering box; c) adjusting the horizontal translation and tilt of the prism of the beam reverser; d) changing the tilt of the output coupler of the powering ring amplifier wavefront engineering box; and e) adjusting the tip or tilt of the fold mirror of the powering ring amplifier wavefront engineering box. Photolithography equipment.

7. 1. An optical element module for use in a photolithography apparatus, comprising: Housing and an optical element positioned within the housing; an actuator extending through a wall of the housing and mechanically coupled to the optical element and adapted to change the position and / or orientation of the optical element in one or more small negative or positive increments that compensate for optical degradation relative to an overall adjustment range for alignment of a position or orientation of the optical element in response to a signal, the components of the actuator for imparting motion to the actuator in response to the signal being located outside the housing; a module arranged to acquire a near-field image of a light beam from a light source, the module having an instrument arranged to extract information from the near-field image; a control system for generating the signal at least partially in response to the information extracted from the near-field image; the optical elements include at least one of a master oscillator output coupler, a master oscillator wavefront engineering box folding mirror, a beam reverser prism, a power ring amplifier wavefront engineering box output coupler, and a power ring amplifier wavefront engineering box folding mirror; The actuator is adapted to at least one of: a) changing the tilt of the output coupler of the master oscillator; b) adjusting the tip or tilt of the folding mirror of the master oscillator wavefront engineering box; c) adjusting the horizontal translation and tilt of the prism of the beam reverser; d) changing the tilt of the output coupler of the powering ring amplifier wavefront engineering box; and e) adjusting the tip or tilt of the folding mirror of the powering ring amplifier wavefront engineering box.

8. the actuator includes a threaded shaft coupled to the optical element; The optical element module of claim 7 , wherein the component of the actuator for imparting motion to the actuator comprises a motor coupled to the threaded shaft and arranged to rotate the threaded shaft.

9. The optical element module of claim 8 , wherein the motor comprises a stepper motor.

10. a light source having at least one optical element; a module arranged to acquire a near-field image of a light beam from the light source, the module having an instrument arranged to extract image information from the near-field image; a control system disposed to receive the image information and for generating a signal at least partially in response to the image information; an actuator mechanically coupled to the at least one optical element, arranged to receive the signal, and adapted in response to the signal to change the position and / or orientation of the at least one optical element in one or more small negative or positive increments that compensate for optical degradation compared to an overall adjustment range for alignment of a position or orientation of the at least one optical element; the at least one optical element comprises at least one of a master oscillator output coupler, a master oscillator wavefront engineering box folding mirror, a beam reverser prism, a power ring amplifier wavefront engineering box output coupler, and a power ring amplifier wavefront engineering box folding mirror; the actuator is adapted to at least one of: a) changing the tilt of the output coupler of the master oscillator; b) adjusting the tip or tilt of the fold mirror of the master oscillator wavefront engineering box; c) adjusting the horizontal translation and tilt of the prism of the beam reverser; d) changing the tilt of the output coupler of the powering ring amplifier wavefront engineering box; and e) adjusting the tip or tilt of the fold mirror of the powering ring amplifier wavefront engineering box. Photolithography equipment.

11. a light source having at least one optical element positioned within a housing; an actuator extending through a wall of the housing and mechanically coupled to the at least one optical element, the actuator adapted to change a position and / or orientation of the at least one optical element in one or more small negative or positive increments that compensate for optical degradation relative to an overall adjustment range for alignment of a position or orientation of the at least one optical element in response to a signal, the actuator having components for imparting motion to the actuator in response to the signal located outside the housing; a combined automatic shutter metrology module arranged to capture an image of a light beam from the light source, the combined automatic shutter metrology module having a metrology instrument arranged to extract information from the image; a control system for generating said signal at least partially in response to said information extracted from said image; the at least one optical element comprises at least one of a master oscillator output coupler, a master oscillator wavefront engineering box folding mirror, a beam reverser prism, a power ring amplifier wavefront engineering box output coupler, and a power ring amplifier wavefront engineering box folding mirror; the actuator is adapted to at least one of: a) changing the tilt of the output coupler of the master oscillator; b) adjusting the tip or tilt of the fold mirror of the master oscillator wavefront engineering box; c) adjusting the horizontal translation and tilt of the prism of the beam reverser; d) changing the tilt of the output coupler of the powering ring amplifier wavefront engineering box; and e) adjusting the tip or tilt of the fold mirror of the powering ring amplifier wavefront engineering box. Photolithography equipment.

12. a light source having at least one optical element positioned within a housing; an actuator extending through a wall of the housing and mechanically coupled to the at least one optical element, the actuator adapted to change a position and / or orientation of the at least one optical element in one or more small negative or positive increments that compensate for optical degradation relative to an overall adjustment range for alignment of a position or orientation of the at least one optical element in response to a signal, the actuator having components for imparting motion to the actuator in response to the signal located outside the housing; a module arranged to obtain a near-field image of a light beam from the light source, the module having an instrument arranged to extract feature information from the near-field image using at least one of a circular Hough transform, a standard Hough transform, and a Radon transform; a control system for generating the signal at least partially in response to the information extracted from the near-field image; the at least one optical element comprises at least one of a master oscillator output coupler, a master oscillator wavefront engineering box folding mirror, a beam reverser prism, a power ring amplifier wavefront engineering box output coupler, and a power ring amplifier wavefront engineering box folding mirror; the actuator is adapted to at least one of: a) changing the tilt of the output coupler of the master oscillator; b) adjusting the tip or tilt of the fold mirror of the master oscillator wavefront engineering box; c) adjusting the horizontal translation and tilt of the prism of the beam reverser; d) changing the tilt of the output coupler of the powering ring amplifier wavefront engineering box; and e) adjusting the tip or tilt of the fold mirror of the powering ring amplifier wavefront engineering box. Photolithography equipment.

Images