DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Acknowledgment is made of applicant’s claim for foreign priority under 35 U.S.C. 119 (a)-(d). The certified copy has been filed in parent Application No. EP24168588.2, filed on 2024.04.04.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim 25 is rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter.
Claim 25 is directed to a "computer readable medium." Applicant describes a computer readable medium by giving an open-ended list on Page 28, Line 23-29 of the specification: " In some implementations, the program code can be stored in one or more machine readable storage devices, such as hard drives, magnetic disks, solid state drives, magneto-optical disks, or optical disks. Alternatively or in addition, the program code can be encoded on a propagated signal that is an artificially generated signal, e.g., a machine-generated electrical, optical, or electromagnetic signal, that is generated to encode information for transmission to suitable receiver apparatus for execution by a programmable processor. " A "computer readable medium" is not explicitly or deliberately defined to include only the non-transitory embodiments listed on Page 28, Line 23-29 The broadest reasonable interpretation of a claim drawn to a computer readable medium typically covers forms of non-transitory tangible media and transitory propagating signals per se in view of the ordinary and customary meaning of computer readable media. See Subject Matter Eligibility of Computer Readable Media, 1351 OG 212 (26 Jan 2010). See MPEP 2111.01. Signals are nothing but the physical characteristics of a form of energy, and as such is nonstatutory natural phenomena. See, e.g., In re Nuitjen, Docket no. 2006-1371 (Fed. Cir. Sept.20, 2007)(slip. op. at 18)("A transitory, propagating signal like Nuitjen's is not a process, machine, manufacture, or composition of matter.' ... Thus, such a signal cannot be patentable subject matter."). Thus, Claim 25 is rejected under 35 U.S.C. (please correct in your template) 101 because, giving the claims their broadest reasonable interpretation, the claimed "computer readable medium" encompasses non-statutory subject matter.
Claims 26 is rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter.
Claim 26 is directed to A computer program product for use in a device which does not fall within at least one of the four categories of patent eligible subject matter recited in 35 U.S.C. 101 (process, machine, manufacture, or composition of matter).
A "computer program product " is not explicitly or deliberately defined to include only the non-transitory embodiments listed on Page 12, Line 10-15. The broadest reasonable interpretation of a claim drawn to a computer readable medium typically covers forms of non-transitory tangible media and transitory propagating signals per se in view of the ordinary and customary meaning of computer readable media. See Subject Matter Eligibility of Computer Readable Media, 1351 OG 212 (26 Jan 2010). See MPEP 2111.01. Signals are nothing but the physical characteristics of a form of energy, and as such is nonstatutory natural phenomena. See, e.g., In re Nuitjen, Docket no. 2006-1371 (Fed. Cir. Sept.20, 2007)(slip. op. at 18)("A transitory, propagating signal like Nuitjen's is not a process, machine, manufacture, or composition of matter.' ... Thus, such a signal cannot be patentable subject matter."). Thus, claims 26 is rejected under 35 U.S.C. 101 because, giving the claims their broadest reasonable interpretation, the claimed "A computer program product for use in a device" is considered to be directed to transitory propagating signals per se and therefore encompasses non-statutory subject matter.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-9, 11-12, 13-18, 20-28 is/are rejected under 35 U.S.C. 103 as being unpatentable over Shi et al. (US 20170309008 A1, hereinafter Shi) in view of Hamouda (WO2024022854A1).
Regarding Claim 1, Shi teaches a computer implemented method for inspecting a photolithography mask to predict defects in wafers, the method comprising:
(Shi, Paragraph [0010], “the lithography model simulates a photolithography process, including the effect of a particular photoresist material”; [0008], "A reticle inspection tool is used to acquire images at different imaging configurations from each of a plurality of pattern areas of a test reticle. A reticle near field for each of the pattern areas of the test reticle is recovered based on the acquired images from each pattern area of the test reticle. A lithography model is applied to the reticle near field for the test reticle to simulate a plurality of test wafer images, and the simulated test wafer images are analyzed to determine whether the test reticle will likely result in an unstable or defective wafer."; Paragraph [0042], "In one embodiment, a reticle qualification is performed by assessing whether the recovered mask near field will likely result in wafer pattern defects under simulated wafer fabrication conditions."); providing a model of the photolithography mask comprising one or more target features and one or more sub-resolution assist features, the photolithography mask being configured for printing of the one or more target features onto a wafer in a printing process using a photolithography system (Shi, Paragraph [0005], "A reticle or photomask is an optical element containing at least transparent and opaque regions, and sometimes semi-transparent and phase shifting regions, which together define the pattern of coplanar features in an electronic device such as an integrated circuit. Reticles are used during photolithography to define specified regions of a semiconductor wafer for etching, ion implantation, or other fabrication processes.", [0059], "Other modifications involve additions of serifs to pattern corners and/or providing nearby sub-resolution assist features (SRAFs), which are not expected to result in printed features and therefore, are referred to as non-printable features."); identifying one or more critical locations in the model of the photolithography mask by verifying a predefined constraint concerning the target features and/or the sub-resolution assist features (Shi, Paragraph [0057], "Thresholds may then be associated with each wafer pattern difference in operation 306. The thresholds may be assigned to different areas of the reticle and, thereby, corresponding wafer patterns. The thresholds may all be the same or be different based on various factors, such as structure type, assigned MEEF (or Mask Error Enhancement Factor as described further below) level or hot spot identification, etc."); generating an aerial image of the photolithography mask comprising the one or more identified critical locations by applying a model of the photolithography system to the photolithography mask (Shi, Paragraph [0008], "A lithography model is applied to the reticle near field for the test reticle to simulate a plurality of test wafer images”, [0054], "After a final calibrated lithography/resist model for a particular process is obtained, such model may be used to generate accurate wafer plane resist images (e.g., after development or after etch) from a mask prior to wafer fabrication with such mask.") ; and predicting defects in wafers by comparing the one or more identified critical locations of the aerial image to one or more corresponding locations of a reference image (Shi, Paragraph [0010], "In one aspect, the lithography model is generated by comparing wafer images resulting from the model with reference images of a wafer that was fabricated using the calibration reticle and adjusting model parameters of the model until a difference between the acquired and reference images is minimized."; [0011], "In another aspect, the simulated test images are analyzed by comparing the simulated test images to images formed from the pre-OPC design database to determine whether the test reticle will likely result in an unstable or defective wafer."; [0010], "In a specific implementation, the model is applied to the test reticle near field under different lithography process conditions. In this aspect, analyzing the simulated test wafer images includes determining whether the test reticle will likely result in an unstable wafer under the different lithography process conditions by comparing the simulated test images having different process conditions and that are associated with a same reticle area."; [0010], "In a further aspect, the test reticle is determined to be unstable when comparing the simulated test images results in a difference above a predefined threshold.")
Hamouda further teaches providing a model of the photolithography mask comprising one or more target features and one or more sub-resolution assist features, the photolithography mask being configured for printing of the one or more target features onto a wafer in a printing process using a photolithography system (Hamouda, Paragraph [0023], "In lithography, to print a target pattern (also often referred to as “design layout” or “design” or “target layout”) on a substrate, a pattern of a patterning device (e.g., a “mask pattern” of a mask) is projected onto a layer of resist provided on a substrate (e.g., a wafer).",; [0008] "inputting a target image to a neural network, the target image associated with a target layout to be printed on a substrate, wherein the neural network is configured to receive a reference image, the reference image corresponding to an optical proximity correction (OPC) mask image of the target image; generating, using the neural network, a predicted mask image representing a mask pattern to be used for printing the target layout on a substrate); identifying one or more critical locations in the model of the photolithography mask by verifying a predefined constraint concerning the target features and/or the sub-resolution assist features (Hamouda, Paragraph [0043], "In some embodiments, the MRC specification may include a minimum critical dimension (CD) of the mask feature that can be manufactured, a minimum curvature of mask feature that can be manufactured, a minimum area of a mask feature, a minimum space between two features, or other geometric properties associated with a mask feature. An MRC violation may occur when the geometric properties of the mask feature do not satisfy the constraints specified in the MRC."; "For example, an MRC violation may occur when the area of the mask feature is lesser than the minimum area specified in the MRC. In another example, the MRC violation may occur when the space between two mask features is lesser than the minimum space specified in the MRC."; [0057]"performing an MRC evaluation may involve determining whether geometric properties of a mask feature in the mask pattern complies with the MRC specification. For example, performing MRC may involve determining whether a size of a mask feature is greater than a minimum size specified in the MRC specification."; generating an aerial image of the photolithography mask comprising the one or more identified critical locations by applying a model of the photolithography system to the photolithography mask (Hamouda, Paragraph [0037], "An aerial image 230 can be simulated from the source model 200, the projection optics model 210 and the patterning device / design layout model module 220. An aerial image (Al) is the radiation intensity distribution at substrate level."; [0062], "At process P704, the loss function component 450 simulates a first image 705 from the predicted mask image 415a. The first simulated image 705 may be an aerial image, a resist image or an etch image."); and predicting defects in wafers by comparing the one or more identified critical locations of the aerial image to one or more corresponding locations of a reference image (Hamouda, Paragraph [0008], "computing a loss function that is indicative of (a) a difference between the predicted mask image and the reference image, and (b) at least one of an MRC evaluation of the predicted mask image or an evaluation of a first simulated image of the predicted mask image", [0064], "At process P708, the loss function component 450 determines the first simulation loss 430 as a function of the difference between the first simulated image 705 and the second simulated image 710.");
Hamouda and Shi are analogous since both deal with photolithography mask modeling, inspection, and wafer-defect prediction using lithography simulation and mask features including SRAFs. Shi qualifies a reticle by recovering a mask near field, applying a lithography model to simulate wafer images, and analyzing them (including comparison to reference images and thresholding) to determine instability/defects. Hamouda trains a prediction model to generate a mask image from a target layout, enforcing MRC constraints (minimum space/area/size/shape) and process metrics (e.g., SRAF printing), and simulates aerial/resist/etch images for comparison to reference images. Therefore, it would have been obvious to incorporate Hamouda’s MRC-aware mask modeling and constraint-based critical-location identification into Shi’s framework so that the mask model includes target features and SRAFs, critical locations are identified by verifying predefined geometric/process constraints, and aerial images are generated and compared to reference images to predict defects. The motivation is to ensure mask features (including SRAFs) satisfy manufacturability/process constraints before printing discussed by Hamouda in Paragraph [0024][0043]–[0044],
Regarding Claim 2, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein the predicted defects comprise sub-resolution assist feature related defects, that is defects resulting from sub-resolution assist features
(Shi, Paragraph [0059], "Other modifications involve additions of serifs to pattern corners and/or providing nearby sub-resolution assist features (SRAFs), which are not expected to result in printed features and therefore, are referred to as non-printable features.");
Hamouda further teaches the same limitation (Hamouda, Paragraph [0044], "Similarly, another process related metric such as SRAF printing, which causes certain features in the mask pattern (e.g., an SRAF) that are not intended to be printed on the substrate to be printed, may occur due to incorrect geometry of features (e.g., when a size of SRAF is greater than a threshold size).", [0024]; Hamouda; "The trained prediction model may then be used to predict a mask image in which the MRC violations or other process metrics such as EPE, SRAF printing are minimized or eliminated.";
Hamouda and Shi are analogous in mask modeling and defect prediction involving SRAFs. Shi models masks with SRAFs as non-printable features and evaluates simulated wafer patterns for instability/defects under varying process conditions. Hamouda treats SRAF printing as a process metric minimized via a loss function that evaluates simulated images in regions far from target features to detect unintended SRAF printing. Therefore, it would have been obvious to incorporate Hamouda’s SRAF-printing evaluation into Shi so that predicted defects include SRAF-related defects (SRAFs printing when they should not).The motivation is to prevent unintended assist-feature printing that causes wafer defects which is discussed by Hamouda in Paragraph [0044][0068].
Regarding Claim 3, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein the predefined constraint comprises a space constraint between target features and/or sub-resolution assist features
(Hamouda, Paragraph [0043], "In some embodiments, the MRC specification may include a minimum critical dimension (CD) of the mask feature that can be manufactured, a minimum curvature of mask feature that can be manufactured, a minimum area of a mask feature, a minimum space between two features, or other geometric properties associated with a mask feature."; "An MRC violation may occur when the geometric properties of the mask feature do not satisfy the constraints specified in the MRC. For example, an MRC violation may occur when the area of the mask feature is lesser than the minimum area specified in the MRC. In another example, the MRC violation may occur when the space between two mask features is lesser than the minimum space specified in the MRC.";
Hamouda and Shi are analogous in mask modeling with constraint-based evaluation. Shi assigns thresholds to reticle areas and identifies hot spots based on pattern differences under varying process conditions. [Hamouda defines MRC specifications including a minimum space between features and identifies violations when the space constraint is not satisfied. Therefore, it would have been obvious to incorporate Hamouda’s space constraint into Shi so that the predefined constraint comprises a space constraint between target features and/or SRAFs. The motivation is to ensure manufacturability and avoid spacing-rule violations that cause printing errors which is discussed by Hamouda in Paragraph [0043].
Regarding Claim 4, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein the space constraint comprises a minimum distance between target features and/or sub-resolution assist features (Hamouda, Paragraph [0043], "In some embodiments, the MRC specification may include a minimum critical dimension (CD) of the mask feature that can be manufactured, a minimum curvature of mask feature that can be manufactured, a minimum area of a mask feature, a minimum space <read on minimum distance> between two features, or other geometric properties associated with a mask feature."; Hamouda; "An MRC violation may occur when the space between two mask features is lesser than the minimum space specified in the MRC.").
Hamouda and Shi are analogous in mask modeling with constraint-based evaluation. Shi assigns thresholds to reticle areas and identifies hot spots based on pattern differences. Hamouda defines MRC specifications including a minimum space (minimum distance) between features and flags violations when space is below the minimum. Therefore, it would have been obvious to incorporate Hamouda’s minimum-distance rule into Shi so that the space constraint comprises a minimum distance between target features and/or SRAFs. The motivation is to prevent bridging/printing failures from overly close features discussed by Hamouda in Paragraph [0043].
Regarding Claim 5, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein the predefined constraint comprises an area constraint defining a minimum area of target features and/or sub-resolution assist features (Hamouda, Paragraph [0043], "In some embodiments, the MRC specification may include a minimum critical dimension (CD) of the mask feature that can be manufactured, a minimum curvature of mask feature that can be manufactured, a minimum area of a mask feature, a minimum space between two features, or other geometric properties associated with a mask feature”; "An MRC violation may occur when the area of the mask feature is lesser than the minimum area specified in the MRC.").
Hamouda and Shi are analogous in mask modeling with constraint-based evaluation. Shi assigns thresholds to reticle areas and identifies hot spots based on pattern differences. Hamouda defines MRC specifications including a minimum area of a mask feature and identifies violations when area is below the minimum. Therefore, it would have been obvious to incorporate Hamouda’s area constraint into Shi so that the predefined constraint comprises a minimum area of target features and/or SRAFs. The motivation is to ensure features have sufficient area for reliable manufacture/printing which discussed by Hamouda in Paragraph [0043].
Regarding Claim 6, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein the predefined constraint comprises a size constraint defining a minimum size of one or more dimensions of target features and/or sub-resolution assist features (Hamouda, Paragraph [0043], "a minimum critical dimension (CD) <read on size constraint> of the mask feature that can be manufactured", [0057]; Hamouda; "performing an MRC evaluation may involve determining whether geometric properties of a mask feature in the mask pattern complies with the MRC specification. For example, performing MRC may involve determining whether a size of a mask feature is greater than a minimum size specified in the MRC specification. If the size is lesser than the minimum size, then the loss function component may identify an MRC violation”).
Hamouda and Shi are analogous in mask modeling with constraint-based evaluation. Shi assigns thresholds to reticle areas and identifies hot spots. Hamouda defines MRC specifications including a minimum critical dimension (CD)/size and identifies violations when size is below the minimum. Therefore, it would have been obvious to incorporate Hamouda’s size constraint into Shi so that the predefined constraint comprises a minimum size of one or more dimensions of target features and/or SRAFs. The motivation is to meet minimum dimensional requirements for manufacturability discussed by Hamouda in Paragraph [0043], [0057].
Regarding Claim 7, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein the predefined constraint comprises a predefined shape of a target feature and/or a sub-resolution assist feature (Hamouda, Paragraph [0043],"The mask features may be of any of various shapes, e.g., curvilinear mask feature."; "An MRC violation may occur when the geometric properties of the mask feature do not satisfy the constraints specified in the MRC.", [0044]; "including incorrect geometry of features (e.g., size, shape, position in the mask pattern, etc.) of the mask pattern."; Shape is geometric property; MRC constraint concerning shape taught.)
Hamouda and Shi are analogous in mask modeling with geometric constraints. Shi assigns thresholds to reticle areas and identifies hot spots. Hamouda defines MRC specifications including shape (e.g., curvilinear features) and flags violations when shape constraints are not satisfied. Therefore, it would have been obvious to incorporate Hamouda’s shape constraint into Shi so that the predefined constraint comprises a predefined shape of a target feature and/or SRAF. The motivation is to ensure features conform to allowable shapes for reliable manufacturing which is discussed by Hamouda in Paragraph [0043]–[0044].
Regarding Claim 8, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein the predefined constraint comprises a predefined pattern of target features and/or sub-resolution assist features (Shi, Paragraph [0008], "A reticle inspection tool is used to acquire images at different imaging configurations from each of a plurality of pattern areas of a test reticle.");
Hamouda further teaches the same limitation (Hamouda, Paragraph [0045], "The mask pattern may have a number of mask features."; [0043], "Typically, a mask may have thousands or even millions of mask features for which MRC may be performed.").
Hamouda and Shi are analogous in mask pattern evaluation. Shi acquires images from multiple pattern areas and evaluates them for stability/defects. Hamouda performs MRC evaluation on mask patterns with thousands/millions of features, implying pattern-level constraints. Therefore, it would have been obvious to incorporate Hamouda’s pattern-level MRC into Shi so that the predefined constraint comprises a predefined pattern of target features and/or SRAFs. The motivation is to ensure overall pattern manufacturability which discussed by Hamouda in Paragraph [0043][0045].
Regarding Claim 9, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein the predefined constraint comprises a predefined location of target features and/or sub-resolution assist features" (Shi, Paragraph [0057]; "Thresholds may then be associated with each wafer pattern difference in operation 306. The thresholds may be assigned to different areas of the reticle and, thereby, corresponding wafer patterns.", [0057]);
Hamouda further teaches the same limitation (Hamouda, Paragraph [0044], "including incorrect geometry of features (e.g., size, shape, position in the mask pattern, etc.)", [0064]; "the loss function component 450 may perform the comparison of the simulated images in a region near (e.g., within a specified proximity < read on Location/position constraint > of) a target feature (e.g., features that are intended to be printed on a substrate).").
Hamouda and Shi are analogous in location-based mask evaluation. Shi assigns thresholds to different reticle areas, tying constraints to locations. Hamouda evaluates simulated images near target features and treats position as a geometric property in MRC. Therefore, it would have been obvious to incorporate Hamouda’s location-based evaluation into Shi so that the predefined constraint comprises a predefined location of target features and/or SRAFs. The motivation is to focus evaluation where printing errors are most likely which is discussed by Hamouda in Paragraph [0064].
Regarding Claim 11, the combination of Shi and Hamouda teaches the invention in Claim 9.
The combination further teaches wherein the photolithography mask is associated with a layer of a semiconductor design, and the predefined location comprises a target feature, which overlaps with another target feature, in particular in a layer above or below the layer associated with the photolithography mask (Shi, Paragraph [0050], "Regardless of form, the calibration wafer structures may be printed in a variety of different wafer layers. In particular, the printed structures are generally printed in a layer of photoresist using standard lithography processes (e.g., projecting a circuit image through a reticle and onto a silicon water coated with photoresist). The wafer may be a calibration wafer with layers of materials that correspond to the materials typically present on product wafers at that step in the test process. The printed structures may be printed over other structures in underlying layers.”).
Regarding Claim 12, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein each target feature and each sub-resolution assist feature is associated with a metric value and the predefined constraint is associated with a limit value, and wherein a predefined number of critical locations is selected from the identified one or more critical locations according to the distance of the metric value of the corresponding target feature or sub-resolution assist feature to the limit value (Shi, Paragraph [0010], "In a further aspect, the test reticle is determined to be unstable when comparing the simulated test images results in a difference above a predefined threshold. In yet a further aspect, different reticle areas have different predefined thresholds."; [0009], " the reticle near field is recovered using a quasi-Newton or conjugate gradient technique for determining the reticle near field. In another aspect, the reticle near field is recovered by a regressive technique that minimizes a sum of a plurality of squared differences between the acquired images and images that are calculated from the reticle near field”).
Hamouda further teaches the same limitation ( Hamouda, Paragraph [0058], "At process P606, the loss function component 450 may assign a violation score 610 to each of the MRC violations 605. The violation score 610 may be determined in a number of ways. In some embodiments, the violation score 610 may be determined as a function of the MRC violations 605 . For example, the greater the magnitude <read on distance >of the MRC violation, the greater may be the violation score 610 <read on Metric value>."; [0068], "At process P806, the loss function component 450 compares pixel values of the third simulated image 805 to a threshold value and computes a score 810 as a function of those pixel values exceeding the threshold value < read on limit value >”; it is noted greater magnitude greater score, above threshold unstable. Selecting predefined number according to distance is obvious to try to prioritize worst violations).
Hamouda and Shi are analogous in threshold/metric-based mask evaluation. Shi determines instability when simulated images differ beyond a threshold, with area-specific thresholds. Hamouda assigns violation scores to MRC violations based on magnitude and compares pixel values to thresholds to compute scores. Therefore, it would have been obvious to incorporate Hamouda’s scoring into Shi so that features have metric values, constraints have limit values, and critical locations are selected by distance to the limit. The motivation is to prioritize worst violations for inspection/repair which is discussed by Hamouda in Paragraph [0058][0068].
Regarding Claim 13, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein the photolithography mask is associated with a layer of a semiconductor design, and the aerial image simulates back-scattering of light from printed layers below the layer associated with the photolithography mask (Shi, Paragraph [0050], "The printed structures may be printed over other structures in underlying layers.", [0042]; "The model may include just the effect of the photolithography scanner, and/or it may also include the effect of resist, etch, CMP or any other wafer processes."; it is noted that the lithography model simulates wafer processes including underlying printed layers (multi-layer stack), which inherently accounts for back-scattering effects from those layers during aerial image simulation).
Regarding Claim 14, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein the model of the photolithography mask comprises a CAD model comprising the one or more target features and the one or more sub-resolution assist features (Shi, Paragraph [0011], "In another aspect, the simulated test images are analyzed by comparing the simulated test images to images formed from the pre-OPC design database to determine whether the test reticle will likely result in an unstable or defective wafer.");
Hamouda further teaches the same limitation (Hamouda, Paragraph [0046], "The target layout includes a number of features and is generally defined as a pre-OPC design layout which can be provided in a standardized digital file format such as GDSII or OASIS or other file format <read on CAD models comprising target and SRAF features >"). Hamouda and Shi are analogous in design-data-based mask modeling. Shi compares simulated images to pre-OPC design database images to determine defects. Hamouda defines the target layout as a pre-OPC design in GDSII/OASIS (a CAD model). Therefore, it would have been obvious to incorporate Hamouda’s CAD model into Shi so that the mask model comprises a CAD model with target features and SRAFs. The motivation is to use industry-standard design formats which is discussed by Hamouda in Paragraph [0046].
Regarding Claim 15, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein the reference image comprises a software simulation of an aerial image of the model of the photolithography mask comprising the one or more identified critical locations (Shi, Paragraph [0076], "The detected signals may also take the form of aerial images. That is, an aerial imaging technique may be used to simulate the optical effects of the photolithography system so as to produce an aerial image of the photoresist pattern that is exposed on the water.");
Hamouda further teaches the same limitation (Hamouda, Paragraph [0037], "An aerial image 230 can be simulated from the source model 200, the projection optics model 210 and the patterning device / design layout model module 220. An aerial image (Al) is the radiation intensity distribution at substrate level.", [0062]; Hamouda; "At process P704, the loss function component 450 simulates a first image 705 <read on reference image> from the predicted mask image 415a. The first simulated image 705 may be an aerial image, a resist image or an etch image”).
Hamouda and Shi are analogous in aerial-image simulation. Shi forms aerial images from detected signals to simulate photolithography optics. Hamouda simulates aerial images from source/projection/patterning models and uses them as reference/first simulated images. Therefore, it would have been obvious to incorporate Hamouda’s aerial simulation into Shi so that the reference image is a software-simulated aerial image of the mask model at critical locations. The motivation is high-fidelity reference images without physical wafers which is discussed by Hamouda in Paragraph [0037][0062].
Regarding Claim 16, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein the reference image comprises an aerial image of a photolithography mask obtained from a database, the aerial image comprising target features substantially identical to the target features in the one or more identified critical locations of the photolithography mask (Shi, Paragraph [0011], "In another aspect, the simulated test images are analyzed by comparing the simulated test images to images formed from the pre-OPC design database to determine whether the test reticle will likely result in an unstable or defective wafer."; [0009],"In another example, the reticle near field is recovered without using a design database that was used to fabricate the reticle”);
Hamouda further teaches the same limitation (Hamouda, Paragraph [0008], "inputting a target image to a neural network, the target image associated with a target layout to be printed on a substrate, wherein the neural network is configured to receive a reference image, the reference image corresponding to an optical proximity correction (OPC) mask image of the target image").
Hamouda and Shi are analogous in database/reference-image usage. Shi compares simulated images to pre-OPC database images. Hamouda uses a reference image corresponding to an OPC mask image and simulates aerial/resist/etch images from it. Therefore, it would have been obvious to incorporate Hamouda’s database reference into Shi so that the reference image is an aerial image from a database with target features identical to the critical locations. The motivation is consistent reference images across masks/processes which is discussed by Hamouda Paragraph [0008][0063].
Regarding Claim 17, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein the reference image comprises an aerial image of a second photolithography mask comprising target features substantially identical to the target features in the one or more identified critical locations of the photolithography mask (Shi, Paragraph [0010], "In one aspect, the lithography model is generated by comparing wafer images resulting from the model with reference images of a wafer that was fabricated using the calibration reticle and adjusting model parameters of the model <read on second photolithography mask used as reference > until a difference between the acquired and reference images is minimized.").
Regarding Claim 18, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches wherein the reference image comprises an acquired aerial image or a simulated aerial image of a different portion of the photolithography mask comprising target features substantially identical to the target features in the one or more identified critical locations of the photolithography mask
(Shi, Paragraph [0008], "A reticle inspection tool is used to acquire images at different imaging configurations from each of a plurality of pattern areas of a test reticle.", [0009]; Shi; "images are acquired at different imaging configurations from each of a plurality of pattern areas of a calibration reticle.").
Regarding Claim 20, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches comparing the aerial image to the reference image comprises the computation of at least one distance measure, in particular at least one distance metric. (Shi, Paragraph [0009], " the reticle near field is recovered using a quasi-Newton or conjugate gradient technique for determining the reticle near field. In another aspect, the reticle near field is recovered by a regressive technique that minimizes a sum of a plurality of squared differences between the acquired images and images that are calculated from the reticle near field”. [0010], "In a further aspect, the test reticle is determined to be unstable when comparing the simulated test images results in a difference above a predefined threshold.").
Hamouda teaches the same limitation (Hamouda, Paragraph [0064], "At process P708, the loss function component 450 determines the first simulation loss 430 as a function of the difference between <read on distance measures > the first simulated image 705 and the second simulated image 710.").
Hamouda and Shi are analogous in image-comparison metrics. Shi minimizes squared differences between acquired and calculated images and thresholds differences. Hamouda computes a loss function from differences between simulated images (a distance measure). Therefore, it would have been obvious to incorporate Hamouda’s loss-based difference into Shi so that comparing aerial to reference images uses at least one distance measure (e.g., squared difference/RMS The motivation is consistent, optimization-friendly quantification of image differences which is discussed by Hamouda in Paragraph [0051][0064].
Regarding Claim 21, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches comparing the aerial image to the reference image comprises the application of a machine learning model (Hamouda, Paragraph [0008], "In some embodiments, there is provided a non-transitory computer readable medium having instructions that, when executed by a computer, cause the computer to execute a method for training a machine learning model to generate a mask image to be used for printing a target layout on a substrate.", [0024],"The prediction model is configured to predict a mask image (e.g., of a mask pattern) from a target image (e.g., of a target layout). The prediction model may be trained based on a loss function (or cost function) that considers a first loss function, which is indicative of image reconstruction loss (e.g., a difference between a predicted mask image and a ground truth mask image) and a second loss function which is indicative of at least one of (a) an MRC loss or (b) a simulation loss (e.g., determined based on an evaluation of images simulated from the predicted mask image to determine process metrics such as EPE, SRAF printing, etc.)."
Hamouda and Shi are analogous in predictive modeling for masks. Shi uses regression (quasi-Newton/conjugate gradient) to recover mask near field and simulate wafer images. Hamouda trains/applies a neural network to generate predicted mask images and evaluate simulated images via loss functions. Therefore, it would have been obvious to incorporate Hamouda’s ML model into Shi so that comparing aerial to reference images uses a machine learning model. The motivation is improved accuracy/efficiency via learned models which is discussed by Hamouda in Paragraph [0024][0047].
Regarding Claim 22, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches comprising using the predicted defects to modify the model of the photolithography mask, in particular the sub-resolution assist features of the model of the photolithography mask. (Shi, Paragraph [0059], "As densities and complexities of integrated circuits (ICs) continue to increase, inspecting photolithographic mask patterns become progressively more challenging. Every new generation of ICs has denser and more complex patterns that currently reach and exceed optical limitations of lithographic systems. To overcome these optical limitations, various Resolution Enhancement Techniques (RET), such as Optical Proximity Correction (OPC), have been introduced. For example, OPC helps to overcome some diffraction limitations by modifying photomask patterns such that the resulting printed patterns correspond to the original desired patterns.".
Hamouda teaches the same limitation (Hamouda, Paragraph [0024], "The prediction model may be trained based on a loss function (or cost function) that considers a first loss function, which is indicative of image reconstruction loss (e.g., a difference between a predicted mask image and a ground truth mask image) and a second loss function which is indicative of at least one of (a) an MRC loss or (b) a simulation loss"; "After the image reconstruction loss, and at least one of the MRC loss or simulation loss are determined, the prediction model may be modified (e.g., parameters of the prediction model) such that the loss function is minimized.").
Hamouda and Shi are analogous in defect-driven mask modification. Shi modifies design/repairs/discards reticles based on predicted unstable/defective wafers. Hamouda modifies the prediction model (e.g., neural network parameters) to minimize MRC/process losses (e.g., SRAF printing). Therefore, it would have been obvious to incorporate Hamouda’s model-update approach into Shi so that predicted defects modify the mask model, particularly SRAFsThe motivation is iterative improvement to reduce violations/printing errors which is discussed by Hamouda in Paragraph [0024][0053].
Regarding Claim 23, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches comprising using the predicted defects to repair the photolithography mask. (Shi, Paragraph [0011], "In a further application, the method includes repairing the test reticle, discarding the test reticle, or monitoring particular areas of a wafer that is fabricated with such test reticle based on a determination that the test reticle will likely result in an unstable or defective wafer.").
Regarding Claim 24, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches a computer implemented method for wafer-less process window qualification for a photolithography mask and a photolithography system comprising: (Shi, Paragraph [0054], "After a final calibrated lithography/resist model for a particular process is obtained, such model may be used to generate accurate wafer plane resist images (e.g., after development or after etch) from a mask prior to wafer fabrication with such mask. These resist images will allow one to assess the wafer images for any inspection patterns with high fidelity and through different focus and exposure settings.".) iterating the following steps: (Shi, Paragraph [0056], "In one embodiment, the model may simply be applied to the mask near field using a plurality of different process conditions, such as focus and dose, to assess the reticle design stability under varying process conditions."; Shi teaches applying model using plurality of different process conditions, implying iteration over conditions.) for at least one photolithography process parameter of the photolithography system selecting a photolithography process parameter value from a predefined range of photolithography process parameter values (Shi, Paragraph [0043], "The input for the model and its modeling parameters includes a set of process conditions. That is, the model is configured to simulate different sets of process conditions on the reconstructed near field mask. Each set of process conditions generally corresponds to a set of wafer manufacturing process parameters that characterize, or partially characterize the wafer process for forming a wafer pattern from the mask. For example, a particular setting of focus and exposure can be input to the model.", [0036], "Examples include different focus settings, different illumination directions or patterns, different linear polarization for the entire illumination pupil or different parts of the illumination pupil, different apodization settings to obscure different portions of the collection beam, etc."; Shi teaches selecting focus, dose, illumination etc. values from range of process conditions.);
inspecting the photolithography mask to predict defects in wafers using a computer-implemented method for inspecting a photolithography mask to predict defects in wafers according to claim 1, wherein the aerial image of the photolithography mask is generated based on the at least one selected photolithography process parameter value; and (Shi, Paragraph [0008], "A reticle inspection tool is used to acquire images at different imaging configurations from each of a plurality of pattern areas of a test reticle. A reticle near field for each of the pattern areas of the test reticle is recovered based on the acquired images from each pattern area of the test reticle. A lithography model is applied to the reticle near field for the test reticle to simulate a plurality of test wafer images, and the simulated test wafer images are analyzed to determine whether the test reticle will likely result in an unstable or defective wafer."; [0010],"In a specific implementation, the model is applied to the test reticle near field under different lithography process conditions.".
teaches inspection per Claim 1 with aerial image generated based on selected process parameter value. This is same as Claim 1 mapping above, see comments on Claim 1.) providing a process window for the at least one photolithography process parameter by analyzing the predicted defects for different photolithography process parameter values. (Shi, [0063]-[0064],"In one embodiment, it is determined whether the design pattern results in unacceptable wafer pattern variation under a specified range of process conditions (or process window). It is determined whether there is a significant difference due to process variability. If the difference between differently processed wafer patterns is higher than a corresponding threshold, such wafer patterns may be deemed defective.", [0054], "After a final calibrated lithography/resist model for a particular process is obtained, such model may be used to generate accurate wafer plane resist images (e.g., after development or after etch) from a mask prior to wafer fabrication with such mask. These resist images will allow one to assess the wafer images for any inspection patterns with high fidelity and through different focus and exposure settings.")
Hamouda further teaches a computer implemented method for wafer-less process window qualification for a photolithography mask and a photolithography system comprising: (Hamouda, Paragraph [0027], "Thus, the term “optimizing” and “optimization” as used herein refers to or means a process that identifies one or more values for one or more parameters that provide an improvement, e.g., a local optimum, in at least one relevant metric, compared to an initial set of one or more values for those one or more parameters. “Optimum” and other related terms should be construed accordingly. In an embodiment, optimization steps can be applied iteratively to provide further improvements in one or more metrics.").
As explained in rejection of claim 1, the obviousness for combining of Hamouda into Shi is provided above.
Regarding Claim 25, it recites limitations similar in scope to the limitations of claim 1 and the combination of Shi and Hamouda teaches all the limitations as of Claim 24. And Shi discloses these features can be implemented on a computer-readable medium (Shi, Paragraph [0091], such a system includes program instructions/computer code for performing various operations described herein that can be stored on a computer readable media. Examples of machine-readable media… and hardware devices that are specially configured to store and perform program instructions).
Regarding Claim 26, it recites limitations similar in scope to the limitations of claim 1 and the combination of Shi and Hamouda teaches all the limitations as of Claim 1. And Shi discloses these features can be implemented on a computer program product (Shi, Paragraph [0090], The computer system 573 may be configured (e.g., with programming instructions) to provide a user interface (e.g., a computer screen) for displaying resultant intensity values, images, and other inspection results; [0004], An integrated circuit is typically fabricated from a plurality of reticles. Initially, circuit designers provide circuit pattern data, which describes a particular integrated circuit (IC) design, to a reticle production system; [0091], Because such information and program instructions may be implemented on a specially configured computer system, such a system includes program instructions/computer code for performing various operations described herein).
Regarding Claim 27, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination further teaches a wafer-less process window qualification system comprising: a subsystem for generating an aerial image of the photolithography mask; and (Shi, Paragraph [0076], “The detected signals may also take the form of aerial images <read on aerial image of the photolithography mask>. That is, an aerial imaging technique may be used to simulate the optical effects of the photolithography system so as to produce an aerial image <read on generating an aerial image> of the photoresist pattern that is exposed on the wafer.”; [0008], “A reticle inspection tool <read on subsystem for generating an aerial image> is used to acquire images at different imaging configurations from each of a plurality of pattern areas of a test reticle. A reticle near field for each of the pattern areas of the test reticle is recovered based on the acquired images from each pattern area of the test reticle. A lithography model is applied to the reticle near field for the test reticle to simulate a plurality of test wafer images <read on generating an aerial image of the photolithography mask>…”; [0054], “After a final calibrated lithography/resist model for a particular process is obtained, such model may be used to generate accurate wafer plane resist images <read on aerial image> (e.g., after development or after etch) from a mask prior to wafer fabrication with such mask. These resist images will allow one to assess the wafer images for any inspection patterns with high fidelity and through different focus and exposure settings.”) a data analysis device comprising at least one memory and at least one processor configured to perform the steps of a computer implemented method of claim 24 (Shi, Paragraph [0072], “The inspection system 400 may receive input 402 from a high NA inspection tool or a low NA inspector mimicking a scanner (not shown). The inspection system may also include a data distribution system (e.g., 404 a and 404 b) for distributing the received input 402, an intensity signal (or patch) processing system (e.g., patch processors and reticle qualification system (e.g., 412) <read on data analysis device> for mask near field and wafer recovery, process modelling, etc., a network (e.g., switched network 408) for allowing communication between the inspection system components, an optional mass storage device 416 <read on memory>, and one or more inspection control and/or review stations (e.g., 410) for reviewing the identified hot spots, inspection results, etc. Each processor <read on processor> of the inspection system 400 typically may include one or more microprocessor integrated circuits and may also contain interface and/or memory <read on memory> integrated circuits and may additionally be coupled to one or more shared and/or global memory devices.”; [0088], “The signals captured by each sensor (e.g., 554 a and/or 554 b) can be processed by a computer system 573 <read on data analysis device> or, more generally, by one or more signal processing devices, which may each include an analog-to-digital converter configured to convert analog signals from each sensor into digital signals for processing. The computer system 573 typically has one or more processors <read on processor> coupled to input/output ports, and one or more memories <read on memory> via appropriate buses or other communication mechanisms.”).
Regarding Claim 28, it recites limitations similar in scope to the limitations of Claim 27 and therefore is rejected under the same rationale.
Claim(s) 10 is/are rejected under 35 U.S.C. 103 as being unpatentable over Shi et al. (US 20170309008 A1, hereinafter Shi) in view of Hamouda (WO2024022854A1) as applied to Claim 9 above and further in view of Hsieh et al. (US 20180149963 A1, hereinafter Hsieh).
Regarding Claim 10, the combination of Shi and Hamouda teaches the invention in Claim 9.
The combination does not explicitly disclose but Hsieh teaches wherein the predefined location comprises an alignment feature, which is used to align the photolithography mask to a wafer (Hsieh, Paragraph [0021], “Mask blank defects increase pattern defectivity on wafers during subsequent photolithography operations. Particularly, for EUV photolithography where the mask blanks are complex multilayer structures, mask blanks are inspected in detail for defects before the main pattern is reproduced. Alignment marks at predetermined locations on the mask blank act as reference coordinates to determine and register the defect location by the inspection tools. Once a defect is identified, its coordinates are registered (defect registration) using the alignment marks as reference points”).
Hsieh and Shi are analogous since both of them are dealing with photolithography mask inspection and defect registration using alignment features. Shi provided a way of assigning thresholds to different reticle areas and identifying hot spots based on pattern differences, implicitly using location-based constraints. [0057] Hsieh provided a way of using alignment marks at predetermined locations on a mask blank as reference coordinates to register defect locations and align the mask to a wafer. [0021] Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention was made to incorporate the alignment-feature teaching of Hsieh into the modified invention of Shi such that the predefined location comprises an alignment feature used to align the photolithography mask to a wafer. The motivation is to enable accurate defect registration and mask-to-wafer alignment using alignment marks as reference points, as discussed by Hsieh in Paragraph [0021]. [0021].
Claim(s) 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Shi et al. (US 20170309008 A1, hereinafter Shi) in view of Hamouda (WO2024022854A1) as applied to Claim 9 above and further in view of Choi et al. (“Sparse machine learning methodology and its applications to semiconductor manufacturing processes”, 2020, hereinafter Choi).
Regarding Claim 19, the combination of Shi and Hamouda teaches the invention in Claim 1.
The combination does not explicitly disclose but Choi teaches wherein the reference image is generated by a machine learning model, in particular by an autoencoder (Choi, Page 48, Section 3.4, “we conduct a comparative experiment using a real-life data from a multi-pattern photolithography process in wafer fabrication”; “we employed the relevance vector machine (RVM) (Tipping, 2001), which is a kernel-based machine learning method”; Page 59, Section 4.1, “we aim to extract features from high-dimensional signals of process equipment sensors for VM considering the characteristics of the signals that consist of multiple sub-processes. To do so, we present a new unsupervised deep autoencoder (AE) with the clipping fusion regularization”).
Choi and Shi are analogous since both of them are dealing with photolithography process monitoring and defect prediction using machine learning models. Shi provided a way of comparing simulated test images to reference images (e.g., from pre-OPC database or calibration wafer) to determine defects. [0010][0011] Choi provided a way of using a machine learning model, in particular an autoencoder, to extract features and generate reference representations from photolithography process data for virtual metrology. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention was made to incorporate the autoencoder-based reference-image generation of Choi into the modified invention of Shi such that the reference image is generated by a machine learning model, in particular by an autoencoder. The motivation is to leverage unsupervised deep learning for feature extraction and reference generation in semiconductor manufacturing processes, as discussed by Choi in Section 4.1 (Page 59).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US 20160162626 A1 LITHOGRAPHY PROCESS WINDOW PREDICTION BASED ON DESIGN DATA
US 20150346610 A1 AERIAL MASK INSPECTION BASED WEAK POINT ANALYSIS
US 20110299758 A1 Wafer Plane Detection of Lithographically Significant Contamination Photomask Defects
US 20110299759 A1 RETICLE DEFECT INSPECTION WITH MODEL-BASED THIN LINE APPROACHES
US 20090310136 A1 Method for Detection of Oversized Sub-Resolution Assist Features
US 20080301620 A1 SYSTEM AND METHOD FOR MODEL-BASED SUB-RESOLUTION ASSIST FEATURE GENERATION
US 20080247632 A1 Method for Mask Inspection for Mask Design and Mask Production
US 20040009416 A1 Qualifying patterns, patterning processes, or patterning apparatus in the fabrication of microlithographic patterns
Any inquiry concerning this communication or earlier communications from the examiner should be directed to YUJANG TSWEI whose telephone number is (571)272-6669. The examiner can normally be reached 8:30am-5:30pm EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Kent Chang can be reached on (571) 272-7667. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/YuJang Tswei/Primary Examiner, Art Unit 2614
Prosecution Insights