DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Responsive to the communication dated 02/10/2026
Claims 16-20 and 22-36 are presented for examination
Finality
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Response to Arguments -112
Applicant’s arguments, see Page 8, filed 02/10/2026, with respect to the rejection of claims 25-36 under 112 have been fully considered and are persuasive. The rejection of claims 25-36 under 112 has been withdrawn.
Response to Arguments -101
Applicant's arguments filed 02/10/2026 have been fully considered but they are not persuasive.
Applicant argues that the judicial exceptions are successfully integrated into a practical application.
Examiner responds by explaining that, firstly, even in view of Desjardins, the claims fail to provide a practical application. Only additional elements are capable of integrating judicial exceptions into a practical application. The two main cruxes of the invention, the selection of gauges and the development of a sampling plan/adjustment of a model configuration, are both judicial exceptions themselves and therefore cannot be the basis of integration into a practical application. An important distinction to make with the claims is that no sampling using the metrology tool is ever actually performed and usage of the configured simulation system to actually perform a simulation does not appear in the independent claims; the end result is merely the generation of a plan to perform sampling or a setup to perform a simulation. These are two fundamentally different operations with fundamentally different impacts on the eligibility analysis under 101. As an analogy, drawing up blueprints and plans to build a house is clearly a mental process, but actually constructing a house using those plans is clearly not simply a mental process.
While it is noted that there is a potentially stronger argument that configuration of a simulation using the selection of gauges provides integration into a practical application, the fact that it is “locked behind” an ‘or’ with a definitively ineligible claim scope makes argument that the claims as a whole are eligible difficult.
As for the applicant’s arguments that the claims are rejected using an analysis that is “essentially the same as admonished by the new Director in the decision issued in Ex Parte Desjardins, Appeal No. 2024-000567 (PTAB September 26, 2025, Appeals Review Panel Decision) (precedential)” because “The Director there said that: ...under the panel's reasoning, many Al innovations are potentially unpatentable-even if they are adequately described and nonobvious- because the panel essentially equated any machine learning with an unpatentable "algorithm" and the remaining additional elements as "generic computer components," without adequate explanation. Dec. 24. Examiners and panels should not evaluate claims at such a high level of generality,” firstly the claims are not directed to any kind of AI innovation nor is any kind of machine learning recited at all, therefore arguments of alleged misinterpretation of a claimed machine learning system in this application as an unpatentable “algorithm” are moot.
Further, the claim limitations are not analyzed at any higher level of generality than claimed. It should be noted that not evaluating claims at an excessive level of generality is not equivalent to importing limitations from the specification of adding additional detail where such additional detail is not present. Characterizing limitations such as “select a first subset of a plurality of gauges from the set of input gauges based on a first property parameter of the one or more properties;” as mentally selecting a set of gauges from a larger one is not the same as evaluating the claim language at an excessive level of generality.
Applicant argues that the claims are comparable to the claims from McRo and therefore are eligible.
Examiner responds by explaining that, the claims are in no way analogous or comparable to those of McRo.
Firstly, it should be noted that McRo was not found to be eligible because it performs a transformation of data or because it recited rules. McRo was found to be eligible because it enabled the automation of specific animation tasks that previously could only be performed subjectively by humans through a method that fundamentally differed from how the method was performed subjectively.
MPEP §2106.05(a): “For example, in McRO, the court relied on the specification’s explanation of how the particular rules recited in the claim enabled the automation of specific animation tasks that previously could only be performed subjectively by humans, when determining that the claims were directed to improvements in computer animation instead of an abstract idea. McRO, 837 F.3d at 1313-14, 120 USPQ2d at 1100-01.” And MPEP §2106.05(a)(II): “In McRO, the Federal Circuit held claimed methods of automatic lip synchronization and facial expression animation using computer-implemented rules to be patent eligible under 35 U.S.C. 101, because they were not directed to an abstract idea. McRO, 837 F.3d at 1316, 120 USPQ2d at 1103. The basis for the McRO court's decision was that the claims were directed to an improvement in computer animation and thus did not recite a concept similar to previously identified abstract ideas. Id. The court relied on the specification's explanation of how the claimed rules enabled the automation of specific animation tasks that previously could not be automated. 837 F.3d at 1313, 120 USPQ2d at 1101. The McRO court indicated that it was the incorporation of the particular claimed rules in computer animation that "improved [the] existing technological process", unlike cases such as Alice where a computer was merely used as a tool to perform an existing process. 837 F.3d at 1314, 120 USPQ2d at 1102. The McRO court also noted that the claims at issue described a specific way (use of particular rules to set morph weights and transitions through phonemes) to solve the problem of producing accurate and realistic lip synchronization and facial expressions in animated characters, rather than merely claiming the idea of a solution or outcome, and thus were not directed to an abstract idea. 837 F.3d at 1313, 120 USPQ2d at 1101.”
In contrast, MPEP § 2106.04(a)(2)(III)(C): “Another example is FairWarning IP, LLC v. Iatric Sys., Inc., 839 F.3d 1089, 120 USPQ2d 1293 (Fed. Cir. 2016). The patentee in FairWarning claimed a system and method of detecting fraud and/or misuse in a computer environment, in which information regarding accesses of a patient’s personal health information was analyzed according to one of several rules (i.e., related to accesses in excess of a specific volume, accesses during a pre-determined time interval, or accesses by a specific user) to determine if the activity indicates improper access. 839 F.3d. at 1092, 120 USPQ2d at 1294. The court determined that these claims were directed to a mental process of detecting misuse, and that the claimed rules here were "the same questions (though perhaps phrased with different words) that humans in analogous situations detecting fraud have asked for decades, if not centuries." 839 F.3d. at 1094-95, 120 USPQ2d at 1296.”
To clarify, see the opinion of FairWarning for its discussion of McRo. “… The claimed rules in McRO transformed a traditionally subjective process performed by human artists into a mathematically automated process executed on computers. [2016 BL 297537], 2016 U.S. App. LEXIS 16703, [WL] at *8-9. Indeed, Defendants conceded that prior animating processes were "driven by subjective determinations rather than specific, limited mathematical rules," such as the mathematical rules articulated in McRO's claimed method. [2016 BL 297537], 2016 U.S. App. LEXIS 16703, [WL] at *8. Thus, the traditional process and newly claimed method stood in contrast: while both produced a similar result, i.e., realistic animations of facial movements accompanying speech, the two practices produced those results in fundamentally different ways…”
With this in mind, it is firstly unclear how the present invention does such selection and plan generation/model configuration tasks in a way that is fundamentally different from how such selection and generation/configuration would be performed manually nor that such selection or plan generation/model configuration was previously impossible to automate. For example, the Bruguier reference utilized in the art rejection clearly describes using such selection to configure a simulation several years prior to the filing date of the present application ([Col 17 line 38-42] “FIG. 10 schematically depicts an exemplary lithographic projection apparatus whose performance could be simulated and/or optimized utilizing the computational lithography models that are calibrated using the test pattern selection process of present invention.”)
Further, it is unclear how the claimed selection technique is equivalent to the rules recited in McRo. The rules of McRo describe how certain modelling techniques should be applied to a 3D character model based on specific sets of triggers given dialogue. In contrast, the claimed selection merely consists of the selection of two sets of items and combining those sets into a third set, removing any duplicates.
Applicant argues that the human mind could not keep track of 100000 gauge patterns.
Examiner responds by explaining that, firstly, just because something may be complex or take a long time, this does not mean that it is not a mental process. For example, writing a novel that is thousands of pages long may take a long time and be significantly complex, but this does not mean that writing a novel is something that cannot be performed in the human mind.
Further, in each of the claims the actual management of the set of gauges, such as selection of subsets, happens within a computer environment, with the ability to handle so much data merely being a result of instructing a general purpose computer to apply the above mental process. Use of a calculator is an effective analogy: a calculator can calculate complex roots faster than a human mind and would be less likely to make errors with such calculations, but this does not mean that taking the square or cube root of a number is not a mathematic calculation.
(MPEP 2106.05(f)(2): Whether the claim invokes computers or other machinery merely as a tool to perform an existing process. Use of a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application or provide significantly more. See Affinity Labs v. DirecTV, 838 F.3d 1253, 1262, 120 USPQ2d 1201, 1207 (Fed. Cir. 2016) (cellular telephone); TLI Communications LLC v. AV Auto, LLC, 823 F.3d 607, 613, 118 USPQ2d 1744, 1748 (Fed. Cir. 2016) (computer server and telephone unit). Similarly, "claiming the improved speed or efficiency inherent with applying the abstract idea on a computer" does not integrate a judicial exception into a practical application or provide an inventive concept. Intellectual Ventures I LLC v. Capital One Bank (USA), 792 F.3d 1363, 1367, 115 USPQ2d 1636, 1639 (Fed. Cir. 2015).)
Response to Arguments -103
Applicant's arguments filed 02/10/2026 have been fully considered but they are not persuasive.
Applicant argues that no prior art teaches determining a plurality of subsets of gauges.
Examiner responds by explaining that this is taught by the combination of the previously cited references.
In particular, Bruguier teaches selecting a first subset of a plurality of gauges from the set of input gauges based on a first property parameter of the one or more properties ([Col 6 line 8-23] “Embodiments of the present invention describes a method of selecting a subset of test patterns from an initial larger set of test patterns for calibrating a computational lithography model, the method comprising: generating an information matrix for the initial larger set of test patterns, wherein the terms of the information matrix comprise one or more identified model parameters that represent a lithographic process response; and, executing a selection algorithm using terms of the information matrix to select the subset of test patterns that effectively determines values of the identified model parameters that contribute significantly in the lithographic process response, wherein the subset of test patterns characteristically represents the initial larger set of test patterns. The selection algorithm explores coverage relationships existing in the initial larger set of test patterns. “[Col 9 line 8-12] “A general aspect of the invention is to select an optimal subset of test patterns (containing K′ no. of test patterns) in step 304 from a larger set of test patterns (containing K no. of test patterns, where K>K′) as described in more detail below” [Col 10 line 44- 49] “While step 402 is known in the art, the unique characteristic of the present invention is that the behavior of the solver is entirely determined by the input values of the parameters that are fed to the solver. In other words, a gauge can be entirely described by its corresponding input values in a general way, regardless of its geometry.”)
Further, Bruguier also teaches the generation of groups of test gauges (i.e. creation of subsets of the larger set of gauges) by grouping test gauges that have similar sensitivity to different model parameters (i.e. a first group with a particular sensitivity to parameter A, a second group with a particular sensitivity to parameter B, etc.) ([Col 9 line 42-55] “the test patterns are selected such that a selected test pattern is very sensitive to one or more specific model parameters, i.e., small changes in the parameters should be able to induce observable changes in the wafer CD for the test pattern. The test patterns are further selected such that the effect of different model parameters can be clearly distinguished. Test patterns with similar sensitivity to the model parameters are identified, grouped and selected such that no unnecessary duplication of test patterns is retained in the selection results. By achieving the selection above, the smallest possible set of test patterns is identified that achieves high sensitivity to the individual parameters as well as clear distinction between the contributions from different model parameters.”) A group of test gauges taken from the larger set is by definition a subset. It should be noted that the way a final “smallest possible set of test patterns” is generated from the groups of patterns in a way that “no unnecessary duplication of test patterns is retained in the selection results” strongly suggests the mechanism found later in the claim of merging the subsets and selecting non-duplicates from the merged subsets to generate a third subset, however Jayaram is instead relied upon due to its explicit description of such merging and duplicate-removal steps.
Applicant argues that no prior art teaches obtain a finite set of input gauges having one or more properties associated with a patterning process to physically form structures, the finite set having at least 100,000 input gauges and each input gauge corresponding to a pattern physically printed or configured for physical printing on a semiconductor substrate by the patterning process to form structures; select a first subset of a plurality of gauges from the set of input gauges based on a first property parameter of the one or more properties; determine a second subset, different from the first subset, of a plurality of gauges from the set of input gauges based on a second property parameter, different from the first property parameter, of the one or more properties;
Examiner responds by explaining that these features are taught by the previously cited references in combination with new reference Badger (US 20070174012 A1).
In particular, Bruguier teaches obtain a finite set of input gauges having one or more properties associated with a patterning process to physically form structures, the finite set having configured for physical printing on a semiconductor substrate by the patterning process to form structures; ([Col 6 line 8-23] “Embodiments of the present invention describes a method of selecting a subset of test patterns from an initial larger set of test patterns for calibrating a computational lithography model, the method comprising: generating an information matrix for the initial larger set of test patterns, wherein the terms of the information matrix comprise one or more identified model parameters that represent a lithographic process response; and, executing a selection algorithm using terms of the information matrix to select the subset of test patterns that effectively determines values of the identified model parameters that contribute significantly in the lithographic process response, wherein the subset of test patterns characteristically represents the initial larger set of test patterns. The selection algorithm explores coverage relationships existing in the initial larger set of test patterns. “[Col 9 line 8-12] “A general aspect of the invention is to select an optimal subset of test patterns (containing K′ no. of test patterns) in step 304 from a larger set of test patterns (containing K no. of test patterns, where K>K′) as described in more detail below” [Col 10 line 44- 49] “While step 402 is known in the art, the unique characteristic of the present invention is that the behavior of the solver is entirely determined by the input values of the parameters that are fed to the solver. In other words, a gauge can be entirely described by its corresponding input values in a general way, regardless of its geometry.”)
Further it is very clear that the patterns described in Bruguier refer to patterns “physically printed or configured for physical printing on a semiconductor substrate by the patterning process to form structures;” ([Col 2 line 42-49] “As noted, microlithography is a central step in the manufacturing of semiconductor integrated circuits, where patterns formed on semiconductor wafer substrates define the functional elements of semiconductor devices, such as microprocessors, memory chips etc. Similar lithographic techniques are also used in the formation of flat panel displays, micro-electro mechanical systems (MEMS) and other devices.” [Col 4 line 8-18] “Currently, calibration is done using a certain number of 1-dimensional and/or 2-dimensional gauge patterns with wafer measurements. More specifically, those 1-dimensional gauge patterns include, but are not limited to, line-space patterns with varying pitch and CD, isolated lines, multiple lines, etc. and the 2-dimensional gauge patterns typically include line-ends, contacts, and randomly selected SRAM (Static Random Access Memory) patterns. Those skilled in the arts will understand that the present invention is generic enough to accommodate any type of pattern.” [Col 5 line 46-55] “The present invention provides a number of innovations in the area of test pattern selection for model calibration that address the requirements mentioned above, among others. A significant advantage of the present invention is that it provides an improved way to measure characteristics of a given test pattern, and at the same time, provides an efficient way to select a subset of test patterns that adequately represent intended lithographic responses. The terms “calibration test pattern”, “test pattern” and “gauge” are used interchangeably.”)
select a first subset of a plurality of gauges from the set of input gauges based on a first property parameter of the one or more properties; ([Col 6 line 8-23] “Embodiments of the present invention describes a method of selecting a subset of test patterns from an initial larger set of test patterns for calibrating a computational lithography model, the method comprising: generating an information matrix for the initial larger set of test patterns, wherein the terms of the information matrix comprise one or more identified model parameters that represent a lithographic process response; and, executing a selection algorithm using terms of the information matrix to select the subset of test patterns that effectively determines values of the identified model parameters that contribute significantly in the lithographic process response, wherein the subset of test patterns characteristically represents the initial larger set of test patterns. The selection algorithm explores coverage relationships existing in the initial larger set of test patterns. “[Col 9 line 8-12] “A general aspect of the invention is to select an optimal subset of test patterns (containing K′ no. of test patterns) in step 304 from a larger set of test patterns (containing K no. of test patterns, where K>K′) as described in more detail below” [Col 10 line 44- 49] “While step 402 is known in the art, the unique characteristic of the present invention is that the behavior of the solver is entirely determined by the input values of the parameters that are fed to the solver. In other words, a gauge can be entirely described by its corresponding input values in a general way, regardless of its geometry.”)
determine a second subset, different from the first subset, of a plurality of gauges from the set of input gauges based on a second property parameter, different from the first property parameter, of the one or more properties; ([Col 9 line 42-55] “the test patterns are selected such that a selected test pattern is very sensitive to one or more specific model parameters, i.e., small changes in the parameters should be able to induce observable changes in the wafer CD for the test pattern. The test patterns are further selected such that the effect of different model parameters can be clearly distinguished. Test patterns with similar sensitivity to the model parameters are identified, grouped and selected such that no unnecessary duplication of test patterns is retained in the selection results. By achieving the selection above, the smallest possible set of test patterns is identified that achieves high sensitivity to the individual parameters as well as clear distinction between the contributions from different model parameters.” [Examiner’s note: creating groups of test patterns based on which parameters each test pattern is most sensitive to (i.e. a first group with a particular sensitivity to parameter A, a second group with a particular sensitivity to parameter B, etc.) is equivalent to creating different subsets each based on different parameters])
While neither Bruguier or Jarayam explicitly teach the use of “at least 100,000 input gauges,” this is taught by new reference Badger.
In particular, Badger makes obvious at least 100,000 input gauges ([Par 27] “] The method for determining photomask inspection capabilities of the present invention includes first a method for generating test patterns, and a systematic method for arranging them on a mask data set. The method described will permit the generation of numerous, even millions, of different patterns in one mask” [Par 39] “As shown in FIG. 6, sub-arrays 26, each of about 1 mm .times.1 mm in dimension, for example, are arranged in a 9.times.10 array to form a main array 24. A plurality of main arrays 24, each of a size of about 9 mm .times.10 mm, are laid out in a 3.times.3 arrangement 22 which may contain about one million of the test patterns or shapes. Photomask 20 of the type used in lithographic production may have transferred thereon identical or different main arrays 22, in any other number, more or less, as desired.”)
Badger is analogous art because it is within the field of lithographic test pattern processing. It would have been obvious to one of ordinary skill in the art to combine Badger with Bruguier and Jayaram before the effective filing date. One of ordinary skill in the art would have been motivated to make this combination in order to generate test patterns in a way that more completely explores the possible design space, ultimately making any processing based on these patterns more accurate to reality and complete. Badger nodes how previous methods of generating test patterns suffer from too much uniformity, limiting the amount of design space that is explorable by systems processing the test patterns ([Par 7] “One way to identify inspection limits is to design test patterns, vary the dimensions of these test patterns, build a mask containing the patterns and then determine empirically which of these shapes passes mask inspection, as described in the U.S. Pat. Nos. 6,482,557 and 6,721,695. However, most of the patterns used in this approach are varied in an overly simple way across a pattern set. For example, the test patterns or shapes vary in one or two critical dimensions. This is due in part to the difficulties associated with mask inspection. When shapes that do not pass mask inspection are encountered, the tool flags the location. If too many of these flagged locations are present, the mask inspection is not completed. A large number of flagged inspection stops can be difficult to analyze efficiently, rendering the mask effectively uninspectable.”) To this end, Badger presents a method that introduces much deeper variety into the generated set of test patterns ([Par 29] “The flow chart of FIG. 1 provides an overview of the preferred method of the present invention. Test patterns or shapes are generated 10 that would include ordered variations of one or more shape variable, and the patterns are then formed and arranged on a mask 12 so as provide an opportunity to test variables or conditions of interest. The layout of the individual test patterns or shapes is important in practicing the method of the invention. The test patterns or shapes are systematically grouped by rule type with patterns of varying complexity, where complexity is measured by the number of variables that are modified within the patterns. [Par 37] “Four or more variables may also be tested, for example, by combining shape 30b with shape 30c. In this case of four variables, the variables would be nub width a, nub right side length b, nub left side length c, and nub distance to corner d. While the variations in nub dimensions a, b and c would be set out in the manner described in connection with basic arrays 28b. 28b' and 28b'', the dimension of the fourth variable d would be varied in the major row position, from a larger value to a smaller value.”) Overall, one of ordinary skill would have recognized that combining Badger with Bruguier and Jayaram would result in a system capable of generating a more varied initial set of input gauges/test patterns, leading to a more complete design space and ultimately enabling more accurately chosen sets of gauges to be selected and thereby more accurate calibrations.
Applicant argues that no prior art teaches “determine one of more candidate models from the plurality of models.”
Examiner responds by firstly explaining that a plurality of models is indeed taught by the previously cited references. Particularly, Cao teaches a plurality of models ([Page 6 par 4] “Once the model MP R is tuned for each of the mask writers, creating "n" models MPRI ..... MPRn (where n is the number of mask writers), in Step 50, each of the mask writers is tuned from the nominal parameter values that were utilized to generate the initial wafer data MD_1 .... MD n utilizing the parameter values of the reference model MP_R and the adjusted model parameters MPRi… The resulting parameters, Pi, are then utilized to tune the corresponding mask writer (i),”) as well as that models are associated with error values ([Page 6 Par 1] “It is noted that the foregoing model has achieved accuracy of 3σ <3 nm (mask scale) for 1D patterns on several 65 nm and 45 nm masks manufactured with different processes. On 2D patterns, model error 3σ is typically about 10 nm for both calibration and prediction.”) Although Cao does not word-for-word include the phrase “each model of the plurality of models being associated with a model error value after calibration;” the language found in [Page 6 Par 4] (“Once the model MP R is tuned for each of the mask writers, creating "n" models MPRI ..... MPRn (where n is the number of mask writers), in Step 50, each of the mask writers is tuned from the nominal parameter values that were utilized to generate the initial wafer data MD_1 .... MD n utilizing the parameter values of the reference model MP_R and the adjusted model parameters MPRi… The resulting parameters, Pi, are then utilized to tune the corresponding mask writer (i),”) and [Page 5 Par 4] (“As is known, during the calibration process, which is an iterative process, the non-tunable parameters are fixed and the tunable parameters are adjusted until the mask generated by the model (i.e., the simulated mask result) matches the actual mask result produced by the reference mask writer. Thus, the model parameters MP R are adjusted (i.e., calibrated) such that the mask results produced by the model equal the actual mask data associated with the reference mask writer MD R within some predefined error criteria or the best match possible.) clearly describes the calibration of a number of models, with [Page 6 Par 4] describing the generation and tuning of parameters for a number of models and [Page 5 Par 4] giving further details of the calibration process itself.
Further, the claim does not require a “technique to choose a model from among a plurality of models.” The wording of the claim, particularly “determine one or more candidate models from the plurality of models …” is not limited to the selection of a set of pre-calibrated models and does not exclude “determining” such a candidate model by tuning a model already in that plurality.
The passage of [Page 5 Par 4] describes determining a model (as noted in [Page 6 Par 4] this model is amongst a plurality of models) by comparing its error to a reference model. Further, [Page 5 Par 4] explains that while the reference data can be taken from a physical reference mask writer, the reference data can also be taken from other simulated models/mask writers ([Page 5 Par 4] “It is noted that any suitable model for simulating the mask writing performance or a mask writer unit may be utilized in this process. It is further noted that any one of the mask writers is tuned and associated mask data, MD 1 ... MD-n, may be utilized to calibrate the model (i.e., may be utilized as the referenced mask writer).”) This clearly reads on the claim language.
Further, the error described in [Page 6 Par 1] of Cao is clearly a post-calibration error, with the prediction error being the error post-calibration when attempting to actually predict results. ([Page 6 Par 1] “It is noted that the foregoing model has achieved accuracy of 3σ <3 nm (mask scale) for 1D patterns on several 65 nm and 45 nm masks manufactured with different processes. On 2D patterns, model error 3σ is typically about 10 nm for both calibration and prediction.”)
Claim Objections
Claims 22-24 objected to because of the following informalities:
Claims 22-24 recite “the process model.” It is clear that this is meant to refer to the “process computer model” introduced in claim 16, and should be amended to reflect this language to avoid any potential issues with antecedent basis.
Appropriate correction is required.
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.
Claims 16-20 and 22-36 are rejected under 35 U.S.C. 101 because they are directed to an abstract idea without significantly more.
Claim 16 (Statutory Category – Product)
Step 2A – Prong 1: Judicial Exception Recited?
Yes, the claim recites a mental process, specifically:
select a first subset of a plurality of gauges from the set of input gauges based on a first property parameter of the one or more properties; determine a second subset, different from the first subset, of a plurality of gauges from the set of input gauges based on a second property parameter, different from the first property parameter, of the one or more properties;
Determining a subset of a group that has certain properties or parameters and then selecting that subset is practical to perform in the human mind through a combination of observation and evaluation. For example, a person could observe a variety of shapes drawn on a piece of paper with a pencil. To select a subset that has the property of having greater than 4 sides, the person could indicate that each pentagon, octagon etc. is “selected,” as by drawing a circle around each pentagon, octagon etc., creating a list of each pentagon, octagon etc., or otherwise taking note of which elements are selected. This process could then be repeated to select another subset, say a subset of shapes that have the property of being the color green, by indicating the selection of all green shapes on the paper. Performing this process for gauges rather than drawn shapes only differs in what is being observed and what properties are filtered for, the mental process itself is equivalent.
merge the gauges of first subset and of the second subset of gauges to be a merged subset of gauges; determine if the merged subset of gauges include duplicate gauges; select a third subset of gauges from the merged subset of gauges such that the third subset of gauges does not include the duplicate gauges; and
Merging these subsets and then removing the duplicates is a mental process that is practical to perform in the human mind. For example, upon observing a subset containing elements { Red Octagon A, Yellow Pentagon B, Green Pentagon C, Green Octagon D} and a second subset containing elements {Green Pentagon C, Green Octagon D, Green Triangle E, Green Square F}, a person could combine the two subsets, either entirely in their mind or using a pencil and paper to keep track, into a new combined subset { Red Octagon A, Yellow Pentagon B, Green Pentagon C, Green Octagon D, Green Pentagon C, Green Octagon D, Green Triangle E, Green Square F}. The person could then mentally examine the merged subset to and judge which elements are repeated, in this case { Green Pentagon C} and {Green Octagon D} appear twice in the merged subset. With this in mind, the person could remove the repeated elements from the merged subset, as by removing them from the mental selection of indicating their removal on the written record.
Performing this operation with gauges as opposed to generic elements/shapes only differs in what these elements are, the mental process is equivalent.
apply at least the third subset of gauges by use of values from the third subset of gauges to change or configure one or more parameters of a process computer model that simulates an aspect of the patterning process or by use of the reduced number of gauges to be measured in the third subset to produce a sampling plan for a metrology tool to make fewer measurements corresponding to the third subset of gauges than measurement of all of the finite set of input gauges.
Choosing parameters for a system such as this is mental process equivalent to merely making a decision about what the values of the parameters should be. “Applying” the gauge selection to do this is recited at such a high level that it could encompass virtually anything from meticulously analyzing each gauge and mapping each aspect of them to a parameter to simply looking at the set of gauges, mentally concluding that “there are a lot of them” and setting the parameters to arbitrarily chosen values.
Similarly, producing a sampling plan by using the reduced number of gauges in the merged subset is also a mental process that could encompass virtually anything, from meticulously examining the features of each selected gauge, recognizing correlations and patterns, and using these recognized patterns to come up with a plan for future sampling, to simply looking at the set of gauges, mentally concluding that “there are a lot of them,” and deciding that from now on the sampling plan is obtain exactly 0 samples, realizing the goal of producing a sampling plan that “make(s) fewer measurements … than measuring all of the finite set of input gauges.”
Step 2A – Prong 2: Integrated into a Practical Solution?
Insignificant Extra-Solution Activity (MPEP 2106.05(g)) has found mere data gathering and post solution activity to be insignificant extra-solution activity.
Data gathering:
obtain a finite set of input gauges having one or more properties associated with a patterning process to physically form structures, the finite set having at least 100,000 input gauges and each input gauge corresponding to a pattern physically printed or configured for physical printing on a semiconductor substrate by the patterning process to form structures;
Obtaining the set of input gauges is merely gathering data related to a variety of gauges.
Step 2B: Claim provides an Inventive Concept?
No, as discussed with respect to Step 2A, the additional limitations are mere data gathering and do not impose any meaningful limits on practicing the abstract idea and therefore the claim does not provide an inventive concept in Step 2B.
Insignificant Extra-Solution Activity (MPEP 2106.05(g)) has found mere data gathering and post solution activity to be insignificant extra-solution activity.
Data gathering:
obtain a finite set of input gauges having one or more properties associated with a patterning process to physically form structures, the finite set having at least 100,000 input gauges and each input gauge corresponding to a pattern physically printed or configured for physical printing on a semiconductor substrate by the patterning process to form structures;
Obtaining the set of input gauges is merely gathering data related to a variety of gauges.
The courts have found that claim elements equivalent to merely gathering data are not indicative of integration into a practical application nor evidence of an Inventive concept (MPEP 2106.05(g)(Mere Data Gathering)(i) Performing clinical tests on individuals to obtain input for an equation, In re Grams, 888 F.2d 835, 839-40; 12 USPQ2d 1824, 1827-28 (Fed. Cir. 1989);)
In addition, the following are also considered as well-understood, routine, and conventional activities, as discussed in MPEP § 2106.05(d):
applying the subset of initial gauges to change or configure a process computer model to more accurately simulate an aspect of the patterning process
Changing options or configuring a process is equivalent to merely changing some parameters or properties in memory. Therefore, the changing or configuring a process is a well-understood, routine, and conventional activity, as evidenced by:
MPEP § 2106.05(d)(II)(iv) Storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015); OIP Techs., 788 F.3d at 1363, 115 USPQ2d at 1092-93;
MPEP § 2106.05(d)(II)(iii) Electronic recordkeeping, Alice Corp. Pty. Ltd. v. CLS Bank Int'l, 573 U.S. 208, 225, 110 USPQ2d 1984 (2014) (creating and maintaining "shadow accounts"); Ultramercial, 772 F.3d at 716, 112 USPQ2d at 1755 (updating an activity log);
Operating System Settings ([Page 1 Par 1-2])
Configuration file – Wikipedia ([Page 1 Par 1-2])
Moreover, the additional computer elements of claim 16 “A computer program product comprising a non-transitory computer readable medium having instructions therein, the instructions, when executed by a computer system, configured to cause the computer system to at least:.. a process computer model.” are rejected for simply applying a general purpose computer. (MPEP 2106.05(f))
Mere Instructions To Apply An Exception (MPEP 2106.05(f)) has found that simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application or provide significantly more.
The additional elements have been considered both individually and as an ordered combination in the consideration of whether they constitute significantly more, and have been determined not to constitute such.
The claim is ineligible.
Claim 25 (Statutory Category – Product)
Step 2A – Prong 1: Judicial Exception Recited?
Yes, the claim recites a mental process, specifically:
determine one or more candidate models from the plurality of models based on a comparison of the model error value associated with each model with respect to a model error value of a particular model in the plurality of models; select a subset of gauges from the initial gauges for the patterning process based on information generated from the one or more candidate models;
Determining these models is a mental process equivalent to observing the set of models, mentally comparing their error values, and judging which models have the lowest error.
The mental process of selecting gauges “based on” the output of the models is recited at such a high level that it could encompass virtually anything, including observing the output of the models, mentally concluding that “there is a lot of data” and then arbitrarily choosing a set of gauges.
apply the subset of gauges by use of values from the subset of gauges to change or configure one or more parameters of a process computer model that simulates an aspect of the patterning process or by use of the reduced number of gauges to be measured in the subset of gauges to produce a sampling plan for a metrology tool to make fewer measurements corresponding to the subset of gauges than measurement of all of the finite set of initial gauges.
Choosing parameters for a system such as this is mental process equivalent to merely making a decision about what the values of the parameters should be. “Applying” the gauge selection to do this is recited at such a high level that it could encompass virtually anything from meticulously analyzing each gauge and mapping each aspect of them to a parameter to simply looking at the set of gauges, mentally concluding that “there are a lot of them” and setting the parameters to arbitrarily chosen values.
Similarly, producing a sampling plan by using the reduced number of gauges in the merged subset is also a mental process that could encompass virtually anything, from meticulously examining the features of each selected gauge, recognizing correlations and patterns, and using these recognized patterns to come up with a plan for future sampling, to simply looking at the set of gauges, mentally concluding that “there are a lot of them,” and deciding that from now on the sampling plan is obtain exactly 0 samples, realizing the goal of producing a sampling plan that “make(s) fewer measurements … than measuring all of the finite set of input gauges.”
the claim also recites a mathematical process, specifically:
MPEP 2106.4(a)(2)(I): “The mathematical concepts grouping is defined as mathematical relationships, mathematical formulas or equations, and mathematical calculations.”
Further, the MPEP recites: “For example, a step of "determining" a variable or number using mathematical methods or "performing" a mathematical operation may also be considered mathematical calculations when the broadest reasonable interpretation of the claim in light of the specification encompasses a mathematical calculation.”
calibrate, via an optimization algorithm using the initial gauges, a plurality of models, each model of the plurality of models being associated with a model error value after calibration.
Algorithms are mathematic concepts, therefore using one to calibrate a model is also a mathematic concept. This is clearly not merely involving the use of math, the performance of the algorithm is the calibration process itself.
Step 2A – Prong 2: Integrated into a Practical Solution?
Insignificant Extra-Solution Activity (MPEP 2106.05(g)) has found mere data gathering and post solution activity to be insignificant extra-solution activity.
Data gathering:
obtain a finite set of initial gauges having one or more properties associated with a patterning process to physically form structures, the finite set having at least 100,000 initial gauges and each initial gauge corresponding to a pattern physically printed or configured for physical printing on a semiconductor substrate by the patterning process to form structures;
Obtaining the set of input gauges is merely gathering data related to a variety of gauges.
Step 2B: Claim provides an Inventive Concept?
No, as discussed with respect to Step 2A, the additional limitations are mere data gathering and do not impose any meaningful limits on practicing the abstract idea and therefore the claim does not provide an inventive concept in Step 2B.
Insignificant Extra-Solution Activity (MPEP 2106.05(g)) has found mere data gathering and post solution activity to be insignificant extra-solution activity.
Data gathering:
obtain a finite set of initial gauges having one or more properties associated with a patterning process to physically form structures, the finite set having at least 100,000 initial gauges and each initial gauge corresponding to a pattern physically printed or configured for physical printing on a semiconductor substrate by the patterning process to form structures;
Obtaining the set of input gauges is merely gathering data related to a variety of gauges.
The courts have found that claim elements equivalent to merely gathering data are not indicative of integration into a practical application nor evidence of an Inventive concept (MPEP 2106.05(g)(Mere Data Gathering)(i) Performing clinical tests on individuals to obtain input for an equation, In re Grams, 888 F.2d 835, 839-40; 12 USPQ2d 1824, 1827-28 (Fed. Cir. 1989);)
In addition, the following are also considered as well-understood, routine, and conventional activities, as discussed in MPEP § 2106.05(d):
apply the subset of gauges to change or configure one or more parameters of a process computer model that simulates an aspect of the patterning process
Changing options or configuring a process is equivalent to merely changing some parameters or properties in memory. Therefore, the changing or configuring a process is a well-understood, routine, and conventional activity, as evidenced by:
MPEP § 2106.05(d)(II)(iv) Storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015); OIP Techs., 788 F.3d at 1363, 115 USPQ2d at 1092-93;
MPEP § 2106.05(d)(II)(iii) Electronic recordkeeping, Alice Corp. Pty. Ltd. v. CLS Bank Int'l, 573 U.S. 208, 225, 110 USPQ2d 1984 (2014) (creating and maintaining "shadow accounts"); Ultramercial, 772 F.3d at 716, 112 USPQ2d at 1755 (updating an activity log);
Operating System Settings ([Page 1 Par 1-2])
Configuration file – Wikipedia ([Page 1 Par 1-2])
Moreover, the additional computer elements of claim 25 “A computer program product comprising a non-transitory computer readable medium having instructions therein, the instructions, when executed by a computer system, configured to cause the computer system to at least:.. process computer model.” are rejected for simply applying a general purpose computer. (MPEP 2106.05(f))
Mere Instructions To Apply An Exception (MPEP 2106.05(f)) has found that simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application or provide significantly more.
The additional elements have been considered both individually and as an ordered combination in the consideration of whether they constitute significantly more, and have been determined not to constitute such.
The claim is ineligible.
Claim 35 (Statutory Category – Process)
Step 2A – Prong 1: Judicial Exception Recited?
Yes, the claim recites a mental process, specifically:
selecting, by a hardware computer, a first subset of a plurality of gauges from the set of input gauges based on a first property parameter of the one or more properties; determining a second subset, different from the first subset, of a plurality of gauges from the set of input gauges based on a second property parameter, different from the first property parameter, of the one or more properties;
Determining a subset of a group that has certain properties or parameters and then selecting that subset is practical to perform in the human mind through a combination of observation and evaluation. For example, a person could observe a variety of shapes drawn on a piece of paper with a pencil. To select a subset that has the property of having greater than 4 sides, the person could indicate that each pentagon, octagon etc. is “selected,” as by drawing a circle around each pentagon, octagon etc., creating a list of each pentagon, octagon etc., or otherwise taking note of which elements are selected. This process could then be repeated to select another subset, say a subset of shapes that have the property of being the color green, by indicating the selection of all green shapes on the paper. Performing this process for gauges rather than drawn shapes only differs in what is being observed and what properties are filtered for, the mental process itself is equivalent.
Additionally, performing the selection “by a hardware computer” amounts to more than mere instructions to apply the exception using a general-purpose computer.
merging the gauges of first subset and of the second subset of gauges to be a merged subset of gauges; determining if the merged subset of gauges include duplicate gauges; selecting a third subset of gauges from the merged subset of gauges such that the third subset of gauges does not include the duplicate gauges; and
Merging these subsets and then removing the duplicates is a mental process that is practical to perform in the human mind. For example, upon observing a subset containing elements { Red Octagon A, Yellow Pentagon B, Green Pentagon C, Green Octagon D} and a second subset containing elements {Green Pentagon C, Green Octagon D, Green Triangle E, Green Square F}, a person could combine the two subsets, either entirely in their mind or using a pencil and paper to keep track, into a new combined subset { Red Octagon A, Yellow Pentagon B, Green Pentagon C, Green Octagon D, Green Pentagon C, Green Octagon D, Green Triangle E, Green Square F}. The person could then mentally examine the merged subset to and judge which elements are repeated, in this case { Green Pentagon C} and {Green Octagon D} appear twice in the merged subset. With this in mind, the person could remove the repeated elements from the merged subset, as by removing them from the mental selection of indicating their removal on the written record.
Performing this operation with gauges as opposed to generic elements/shapes only differs in what these elements are, the mental process is equivalent.
applying at least the third subset of gauges by use of values from the third subset of gauges to change or configure one or more parameters of a process computer model that simulates an aspect of the patterning process or by use of the reduced number of gauges to be measured in the third subset to produce a sampling plan for a metrology tool to make fewer measurements corresponding to the third subset of gauges than measuring all of the finite set of input gauges.
Choosing parameters for a system such as this is mental process equivalent to merely making a decision about what the values of the parameters should be. “Applying” the gauge selection to do this is recited at such a high level that it could encompass virtually anything from meticulously analyzing each gauge and mapping each aspect of them to a parameter to simply looking at the set of gauges, mentally concluding that “there are a lot of them” and setting the parameters to arbitrarily chosen values.
Similarly, producing a sampling plan by using the reduced number of gauges in the merged subset is also a mental process that could encompass virtually anything, from meticulously examining the features of each selected gauge, recognizing correlations and patterns, and using these recognized patterns to come up with a plan for future sampling, to simply looking at the set of gauges, mentally concluding that “there are a lot of them,” and deciding that from now on the sampling plan is obtain exactly 0 samples, realizing the goal of producing a sampling plan that “make(s) fewer measurements … than measuring all of the finite set of input gauges.”
Step 2A – Prong 2: Integrated into a Practical Solution?
Insignificant Extra-Solution Activity (MPEP 2106.05(g)) has found mere data gathering and post solution activity to be insignificant extra-solution activity.
Data gathering:
obtaining a finite set of input gauges having one or more properties associated with a patterning process to physically form structures, the finite set having at least 100000 input gauges and each input gauge corresponding to a pattern physically printed or configured for physical printing on a semiconductor substrate by the patterning process to form structures;
Obtaining the set of input gauges is merely gathering data related to a variety of gauges.
Step 2B: Claim provides an Inventive Concept?
No, as discussed with respect to Step 2A, the additional limitations are mere data gathering and do not impose any meaningful limits on practicing the abstract idea and therefore the claim does not provide an inventive concept in Step 2B.
Insignificant Extra-Solution Activity (MPEP 2106.05(g)) has found mere data gathering and post solution activity to be insignificant extra-solution activity.
Data gathering:
obtaining a finite set of input gauges having one or more properties associated with a patterning process to physically form structures, the finite set having at least 100000 input gauges and each input gauge corresponding to a pattern physically printed or configured for physical printing on a semiconductor substrate by the patterning process to form structures;
Obtaining the set of input gauges is merely gathering data related to a variety of gauges.
The courts have found that claim elements equivalent to merely gathering data are not indicative of integration into a practical application nor evidence of an Inventive concept (MPEP 2106.05(g)(Mere Data Gathering)(i) Performing clinical tests on individuals to obtain input for an equation, In re Grams, 888 F.2d 835, 839-40; 12 USPQ2d 1824, 1827-28 (Fed. Cir. 1989);)
In addition, the following are also considered as well-understood, routine, and conventional activities, as discussed in MPEP § 2106.05(d):
applying at least the third subset of gauges by use of values from the third subset of gauges to change or configure one or more parameters of a process computer model
Changing options or configuring a process is equivalent to merely changing some parameters or properties in memory. Therefore, the changing or configuring a process is a well-understood, routine, and conventional activity, as evidenced by:
MPEP § 2106.05(d)(II)(iv) Storing and retrieving information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015); OIP Techs., 788 F.3d at 1363, 115 USPQ2d at 1092-93;
MPEP § 2106.05(d)(II)(iii) Electronic recordkeeping, Alice Corp. Pty. Ltd. v. CLS Bank Int'l, 573 U.S. 208, 225, 110 USPQ2d 1984 (2014) (creating and maintaining "shadow accounts"); Ultramercial, 772 F.3d at 716, 112 USPQ2d at 1755 (updating an activity log);
Operating System Settings ([Page 1 Par 1-2])
Configuration file – Wikipedia ([Page 1 Par 1-2])
Moreover, the additional computer elements of claim 16 “A computer program product comprising a non-transitory computer readable medium having instructions therein, the instructions, when executed by a computer system, configured to cause the computer system to at least:... by a hardware computer, a process computer model.” are rejected for simply applying a general purpose computer. (MPEP 2106.05(f))
Mere Instructions To Apply An Exception (MPEP 2106.05(f)) has found that simply adding a general purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not integrate a judicial exception into a practical application or provide significantly more.
The additional elements have been considered both individually and as an ordered combination in the consideration of whether they constitute significantly more, and have been determined not to constitute such.
The claim is ineligible.
Claim 17 recites “wherein the instructions are further configured to cause the computer system to filter the set of the input gauges by use of user defined gauges to determine the first subset of gauges.”
Filtering sets is something that is practical to perform in the human mind, for example, a person could draw a series of squares and circles on a piece of paper. To filter out the circles, the person could look at the paper and erase all the circles, leaving only squares.
Claim 18 recites “wherein the one or more properties comprises one or more selected from: a value of critical dimension of a substrate, a curvature associated with the pattern, and/or an intensity used in the patterning process.
This merely specifies the form that the one or more properties takes, and is therefore merely an extension of the data gathering step
Claim 19 recites “wherein the first property parameter includes a model error, the model error being a difference between a reference contour and a simulated contour generated from a simulation using a process model of the patterning process”
This merely specifies the form that the first property parameter takes, and is therefore merely an extension of the data gathering step
Claim 20 recites “wherein the reference contour is a measured contour from a scanning electron microscope.”
This merely specifies how the reference contour is obtained, and is therefore merely an extension of the data gathering step
Claim 22 recites “wherein the instructions are further configured to cause the computer system to, responsive to a determination that no duplicate gauges exist, select the merged subset of gauges to calibrate the process model.”
Selecting a set is practical to perform in a human mind, equivalent to focusing on something, isolating that thing from a group, or generally deciding which elements a person would like to consider further, as with separating or circling the squares from a collection of many different shapes in a drawing.
Further, performing this process via a computer amounts to no more than mere instructions to apply the exception using a general purpose computer.
Claim 23 recites “wherein the instructions are further configured to cause the computer system to calibrate the process model using the first subset of gauges.
This calibration is done through an algorithm, a type of mathematic process, and is therefore a mathematic concept. See [Par 20] of the specification.
Further, performing this process via a computer amounts to no more than mere instructions to apply the exception using a general purpose computer.
Claim 24 recites “wherein the instructions are further configured to cause the computer system to determine a process condition by simulating using the process model and the selected gauges.”
Simulating the model is a mental process equivalent to drawing a representation of the process step by step, as with a pen and paper. For example, a first step could show an empty processing device, a second step could show an unprocessed wafer prepared on the device, and a third drawn step could show the finished processed wafer. Determining this process condition (note that the spec defines the process condition as [Par 171] “The process condition comprises one or more process parameters, wherein the process parameter is at least one of: dose, focus, or intensity.”) could be as simple as drawing an indication of this condition next to the device, such as a note stating that the device is using a certain dosage.
Further, using a computer to perform this simulation amounts to no more than mere instructions to apply the exception using a general purpose computer.
Applying a computer to perform a generic simulation at a high level of generality is simply the act of instructing a computer to perform generic functions to perform that simulation, which is merely an instruction to apply a computer to the judicial exception. The claim only recites the idea of a solution or outcome, i.e. that the process model is “simulated” without reciting how this simulation is actually accomplished. Further, the computer elements claimed are cited as merely generic tools to perform the operations.
The courts have found that such mere instructions to apply are not indicative of integration into a practical application nor recitation of significantly more than the judicial exception (MPEP 2106.05(f) “Another consideration when determining whether a claim integrates a judicial exception into a practical application in Step 2A Prong Two or recites significantly more than a judicial exception in Step 2B is whether the additional elements amount to more than a recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. As explained by the Supreme Court, in order to make a claim directed to a judicial exception patent-eligible, the additional element or combination of elements must do "‘more than simply stat[e] the [judicial exception] while adding the words ‘apply it’". Alice Corp. v. CLS Bank, 573 U.S. 208, 221, 110 USPQ2d 1976, 1982-83 (2014) (quoting Mayo Collaborative Servs. V. Prometheus Labs., Inc., 566 U.S. 66, 72, 101 USPQ2d 1961, 1965). Thus, for example, claims that amount to nothing more than an instruction to apply the abstract idea using a generic computer do not render an abstract idea eligible. Alice Corp., 573 U.S. at 223, 110 USPQ2d at 1983”)
Claim 26 recites “determine a first subset of gauges from the initial gauges based on a first property of the one or more properties, the first property being a weight and/or a model error; determine a second subset of gauges from the initial gauges based on a second property of the one or more properties; merge the first subset of gauges and the second subset of gauges to be a merged subset of gauges; determine if the merged subset of gauges include duplicate gauges; and selecting a third subset of the merged subset of gauges based on the one or more properties of the patterning process such that the third subset does not include the duplicate gauges.”
Determining a subset of a group that has certain properties or parameters and then selecting that subset is practical to perform in the human mind through a combination of observation and evaluation. For example, a person could observe a variety of shapes drawn on a piece of paper with a pencil. To select a subset that has the property of having greater than 4 sides, the person could indicate that each pentagon, octagon etc. is “selected,” as by drawing a circle around each pentagon, octagon etc., creating a list of each pentagon, octagon etc., or otherwise taking note of which elements are selected. This process could then be repeated to select another subset, say a subset of shapes that have the property of being the color green, by indicating the selection of all green shapes on the paper. Performing this process for gauges rather than drawn shapes only differs in what is being observed and what properties are filtered for, the mental process itself is equivalent.
Merging these subsets and then removing the duplicates is a mental process that is practical to perform in the human mind. For example, upon observing a subset containing elements { Red Octagon A, Yellow Pentagon B, Green Pentagon C, Green Octagon D} and a second subset containing elements {Green Pentagon C, Green Octagon D, Green Triangle E, Green Square F}, a person could combine the two subsets, either entirely in their mind or using a pencil and paper to keep track, into a new combined subset { Red Octagon A, Yellow Pentagon B, Green Pentagon C, Green Octagon D, Green Pentagon C, Green Octagon D, Green Triangle E, Green Square F}. The person could then mentally examine the merged subset to and judge which elements are repeated, in this case { Green Pentagon C} and {Green Octagon D} appear twice in the merged subset. With this in mind, the person could remove the repeated elements from the merged subset, as by removing them from the mental selection of indicating their removal on the written record.
Claim 27 recites “wherein the instructions are further configured to cause the computer system to filter the set of initial gauges by use of user defined gauges to determine the first subset of gauges and the second subset of gauges”
Filtering sets is something that is practical to perform in the human mind, for example, a person could draw a series of squares and circles on a piece of paper. To filter out the circles, the person could look at the paper and erase all the circles, leaving only squares.
Claim 28 recites “ wherein the instructions are further configured to cause the computer system to determine a similarity metric between at least two candidate models.”
Determining the similarity between two things is practical to perform in the human mind, and consists of observing both things and evaluating how similar they are.
Should it be found that this claim is not a mental process, it is also a mathematic process equivalent to calculating the cosine similarity of two vectors representative of the candidate models (see [Par 22] of the specification)
Further, performing this process via a computer amounts to no more than mere instructions to apply the exception using a general purpose computer.
Claim 29 recites “ wherein the similarity metric is a cosine similarity metric being a cosine of two vectors, each vector being representative of a given candidate model of the candidate models.”
Calculating the cosine of two vectors is a mathematic calculation.
Claim 30 recites “wherein the instructions are further configured to cause the computer system to select, based on the similarity metric, a diverse candidate model from the candidate models, wherein the diverse candidate model has a value of the similarity metric substantially different from a value of the similarity metric of a candidate model having least model error value.”
Determining something with low similarity to something else is practical to perform in the human mind, and consists of observing both things and evaluating how different they are.
Selecting a model is a mental process equivalent to an making a decision of what model should be used.
Further, performing this process via a computer amounts to no more than mere instructions to apply the exception using a general purpose computer.
Claim 31 recites “wherein the one or more properties comprise at least one selected from: a value of critical dimension of a substrate, a curvature associated with the pattern, and/or an intensity used in the patterning process.”
This merely specifies the form that the one or more properties are to take, and is therefore merely an extension of the data gathering step.
Claim 32 recites “wherein the model error value associated with each model corresponds to a difference between a reference contour and a simulated contour generated from a simulation using a process model of the patterning process, the reference contour being a measured contour from an image capture device.”
Determining the numerical difference between two things is a mathematic calculation.
Claim 33 recites “wherein the selection of the gauges is based on at least one selected from: a mean value of the model error, a standard deviation value of the model error, and/or a peak-to-peak value of the model”
Determining the mean value of a set of values, the standard deviation of the error values, or the peak-to-peak values are all mathematic calculations
Claim 34 recites “wherein the instructions are further configured to cause the computer system to determine a process condition by simulating using a process model and the selected gauges.”
Simulating the model is a mental process equivalent to drawing a representation of the process step by step, as with a pen and paper. For example, a first step could show an empty processing device, a second step could show an unprocessed wafer prepared on the device, and a third drawn step could show the finished processed wafer. Determining this process condition (note that the spec defines the process condition as [Par 171] “The process condition comprises one or more process parameters, wherein the process parameter is at least one of: dose, focus, or intensity.”) could be as simple as drawing an indication of this condition next to the device, such as a note stating that the device is using a certain dosage.
Further, performing this process via a computer amounts to no more than mere instructions to apply the exception using a general purpose computer.\
Applying a computer to perform a generic simulation at a high level of generality is simply the act of instructing a computer to perform generic functions to perform that simulation, which is merely an instruction to apply a computer to the judicial exception. The claim only recites the idea of a solution or outcome, i.e. that the clinical trial is “simulated” without reciting how this simulation is actually accomplished. Further, the computer elements claimed are cited as merely generic tools to perform the operations; for additional clarity see (SIM ELEMENT EVIDENCE)
The courts have found that such mere instructions to apply are not indicative of integration into a practical application nor recitation of significantly more than the judicial exception (MPEP 2106.05(f) “Another consideration when determining whether a claim integrates a judicial exception into a practical application in Step 2A Prong Two or recites significantly more than a judicial exception in Step 2B is whether the additional elements amount to more than a recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. As explained by the Supreme Court, in order to make a claim directed to a judicial exception patent-eligible, the additional element or combination of elements must do "‘more than simply stat[e] the [judicial exception] while adding the words ‘apply it’". Alice Corp. v. CLS Bank, 573 U.S. 208, 221, 110 USPQ2d 1976, 1982-83 (2014) (quoting Mayo Collaborative Servs. V. Prometheus Labs., Inc., 566 U.S. 66, 72, 101 USPQ2d 1961, 1965). Thus, for example, claims that amount to nothing more than an instruction to apply the abstract idea using a generic computer do not render an abstract idea eligible. Alice Corp., 573 U.S. at 223, 110 USPQ2d at 1983”)
Claim 36 recites “wherein the first property parameter includes a model error, the model error being a difference between a reference contour and a simulated contour generated from a simulation using a process model of the patterning process”
This merely specifies the form that the first property parameter takes, and is therefore merely an extension of the data gathering step
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.
Claims 16-17, 22-23, and 35 are rejected under 35 U.S.C. 103 as being unpatentable over Bruguier (US 9588439 B1) in view of SQL Union overview, usage and examples (hereinafter Jayaram) in further view of Badger (US 20070174012 A1).
Claim 16. Bruguier makes obvious A computer program product comprising a non-transitory computer readable medium having instructions therein, the instructions, when executed by a computer system, configured to cause the computer system to at least: ([Col 15 line 40-50] “FIG. 9 is an exemplary block diagram that illustrates a computer system 100 which can assist in embodying and/or implementing the pattern selection method disclosed herein. Computer system 100 includes a bus 102 or other communication mechanism for communicating information, and one or more processor(s) 104 (and 105) coupled with bus 102 for processing information. Computer system 100 also includes a main memory 106, such as a random access memory (RAM) or other dynamic storage device, coupled to bus 102 for storing information and instructions to be executed by processor 104”) obtain a finite set of input gauges having one or more properties associated with a patterning process to physically form structures, the finite set having substrate by the patterning process to form structures; select a first subset of a plurality of gauges from the set of input gauges based on a first property parameter of the one or more properties; ([Col 6 line 8-23] “Embodiments of the present invention describes a method of selecting a subset of test patterns from an initial larger set of test patterns for calibrating a computational lithography model, the method comprising: generating an information matrix for the initial larger set of test patterns, wherein the terms of the information matrix comprise one or more identified model parameters that represent a lithographic process response; and, executing a selection algorithm using terms of the information matrix to select the subset of test patterns that effectively determines values of the identified model parameters that contribute significantly in the lithographic process response, wherein the subset of test patterns characteristically represents the initial larger set of test patterns. The selection algorithm explores coverage relationships existing in the initial larger set of test patterns. “[Col 9 line 8-12] “A general aspect of the invention is to select an optimal subset of test patterns (containing K′ no. of test patterns) in step 304 from a larger set of test patterns (containing K no. of test patterns, where K>K′) as described in more detail below” [Col 10 line 44- 49] “While step 402 is known in the art, the unique characteristic of the present invention is that the behavior of the solver is entirely determined by the input values of the parameters that are fed to the solver. In other words, a gauge can be entirely described by its corresponding input values in a general way, regardless of its geometry.”) determine a second subset, different from the first subset, of a plurality of gauges from the set of input gauges based on a second property parameter, different from the first property parameter, of the one or more properties; ([Col 9 line 42-55] “the test patterns are selected such that a selected test pattern is very sensitive to one or more specific model parameters, i.e., small changes in the parameters should be able to induce observable changes in the wafer CD for the test pattern. The test patterns are further selected such that the effect of different model parameters can be clearly distinguished. Test patterns with similar sensitivity to the model parameters are identified, grouped and selected such that no unnecessary duplication of test patterns is retained in the selection results. By achieving the selection above, the smallest possible set of test patterns is identified that achieves high sensitivity to the individual parameters as well as clear distinction between the contributions from different model parameters.” [Examiner’s note: creating groups of test patterns based on which parameters each test pattern is most sensitive to (i.e. a first group with a particular sensitivity to parameter A, a second group with a particular sensitivity to parameter B, etc.) is equivalent to creating different subsets each based on different parameters] [Col 10 line 44- 49] “While step 402 is known in the art, the unique characteristic of the present invention is that the behavior of the solver is entirely determined by the input values of the parameters that are fed to the solver. In other words, a gauge can be entirely described by its corresponding input values in a general way, regardless of its geometry.” [Examiner’s note: several input values are used, i.e. a first property parameter, second properties parameter, etc.]) ([Col 6 line 8-12] “… a subset of test patterns …”) and of the second subset of gauges ([Col 9 line 42-55] “the test patterns are selected such that a selected test pattern is very sensitive to one or more specific model parameters, i.e., small changes in the parameters should be able to induce observable changes in the wafer CD for the test pattern. The test patterns are further selected such that the effect of different model parameters can be clearly distinguished. Test patterns with similar sensitivity to the model parameters are identified, grouped and selected such that no unnecessary duplication of test patterns is retained in the selection results. By achieving the selection above, the smallest possible set of test patterns is identified that achieves high sensitivity to the individual parameters as well as clear distinction between the contributions from different model parameters.” [Col 6 line 8-12] “… a subset of test patterns …”) to be a ([Col 6 line 8-12] “… a subset of test patterns …”)
([Col 6 line 8-12] “… a subset of test patterns …”) ([Col 6 line 8-12] “… a subset of test patterns …”) ([Col 6 line 8-12] “… a subset of test patterns …”) ([Col 6 line 8-12] “… a subset of test patterns …”) ([Col 6 line 8-12] “… a subset of test patterns …”) ([Col 6 line 8-12] “… a subset of test patterns …”) and apply at least the ([Col 17 line 38-42] “FIG. 10 schematically depicts an exemplary lithographic projection apparatus whose performance could be simulated and/or optimized utilizing the computational lithography models that are calibrated using the test pattern selection process of present invention.”) or by use of the reduced number of gauges to be measured in the third subset to produce a sampling plan for a metrology tool to make fewer measurements corresponding to the third subset of gauges than measurement of all of the finite set of input gauges.
Bruguier does not explicitly teach the finite set having at least 100,000 input gauges, merge the data of the first subset and of the second subset to be a merged subset; determine if the merged subset of data include duplicate data; select a third subset of data from the merged subset such that the third subset does not include the duplicate data; the third subset of data
Jayaram makes obvious ([Page 3 Par 1] “The Union operator combines the results of two or more queries into a distinct single result set that includes all the rows that belong to all queries in the Union. In this operation, it combines two more queries and removes the duplicates.”)
PNG
media_image1.png
548
789
media_image1.png
Greyscale
Jayaram is analogous art because it is within the field of data selection. It would have been obvious to one of ordinary skill in the art to combine it with Bruguier before the effective filing date. As noted by Bruguier, the removal of duplicate test patterns is known to improve the model calibration process ([Col 9 line 46-55] “The test patterns are further selected such that the effect of different model parameters can be clearly distinguished. Test patterns with similar sensitivity to the model parameters are identified, grouped and selected such that no unnecessary duplication of test patterns is retained in the selection results. By achieving the selection above, the smallest possible set of test patterns is identified that achieves high sensitivity to the individual parameters as well as clear distinction between the contributions from different model parameters”) and that, generally, processing a smaller set of gauges without losing a significant amount of data (as by removing redundancies) results in a more efficient calibration process ([Col 10 line 4-17] “execute a selection algorithm to select the subset of K′ test patterns from the initial larger set of K patterns, where (K>K′), in a way so that the characteristics of all K gauges are sufficiently covered by the K′ test patterns … After having selected the subset of K′ patterns, the calibration of the lithographic model can proceed using conventional methods, except with fewer gauges (step 306), resulting in a computationally efficient calibration.”) Although Bruguier uses a mathematic method to distill the set of gauges into a simplified form, it is still plausible that duplicate gauges and their representations would remain in the reduced subset. To this end, Jayaram describes the UNION operator of SQL, which combines two datasets into a third dataset while removing the duplicates ([Page 3 Par 1] “The Union operator combines the results of two or more queries into a distinct single result set that includes all the rows that belong to all queries in the Union. In this operation, it combines two more queries and removes the duplicates.”) Overall, one of ordinary skill in the art would have recognized that combining Jayaram with Bruguier would result in a system that needs to process fewer gauges to perform calibration, therefore making the overall calibration process more efficient.
The combination of Bruguier and Jarayam does not explicitly teach the finite set having at least 100,000 input gauges
Badger makes obvious the finite set having at least 100,000 input gauges ([Par 27] “] The method for determining photomask inspection capabilities of the present invention includes first a method for generating test patterns, and a systematic method for arranging them on a mask data set. The method described will permit the generation of numerous, even millions, of different patterns in one mask” [Par 39] “As shown in FIG. 6, sub-arrays 26, each of about 1 mm .times.1 mm in dimension, for example, are arranged in a 9.times.10 array to form a main array 24. A plurality of main arrays 24, each of a size of about 9 mm .times.10 mm, are laid out in a 3.times.3 arrangement 22 which may contain about one million of the test patterns or shapes. Photomask 20 of the type used in lithographic production may have transferred thereon identical or different main arrays 22, in any other number, more or less, as desired.”)
Badger is analogous art because it is within the field of lithographic test pattern processing. It would have been obvious to one of ordinary skill in the art to combine Badger with Bruguier and Jayaram before the effective filing date. One of ordinary skill in the art would have been motivated to make this combination in order to generate test patterns in a way that more completely explores the possible design space, ultimately making any processing based on these patterns more accurate to reality and complete. Badger nodes how previous methods of generating test patterns suffer from too much uniformity, limiting the amount of design space that is explorable by systems processing the test patterns ([Par 7] “One way to identify inspection limits is to design test patterns, vary the dimensions of these test patterns, build a mask containing the patterns and then determine empirically which of these shapes passes mask inspection, as described in the U.S. Pat. Nos. 6,482,557 and 6,721,695. However, most of the patterns used in this approach are varied in an overly simple way across a pattern set. For example, the test patterns or shapes vary in one or two critical dimensions. This is due in part to the difficulties associated with mask inspection. When shapes that do not pass mask inspection are encountered, the tool flags the location. If too many of these flagged locations are present, the mask inspection is not completed. A large number of flagged inspection stops can be difficult to analyze efficiently, rendering the mask effectively uninspectable.”) To this end, Badger presents a method that introduces much deeper variety into the generated set of test patterns ([Par 29] “The flow chart of FIG. 1 provides an overview of the preferred method of the present invention. Test patterns or shapes are generated 10 that would include ordered variations of one or more shape variable, and the patterns are then formed and arranged on a mask 12 so as provide an opportunity to test variables or conditions of interest. The layout of the individual test patterns or shapes is important in practicing the method of the invention. The test patterns or shapes are systematically grouped by rule type with patterns of varying complexity, where complexity is measured by the number of variables that are modified within the patterns. [Par 37] “Four or more variables may also be tested, for example, by combining shape 30b with shape 30c. In this case of four variables, the variables would be nub width a, nub right side length b, nub left side length c, and nub distance to corner d. While the variations in nub dimensions a, b and c would be set out in the manner described in connection with basic arrays 28b. 28b' and 28b'', the dimension of the fourth variable d would be varied in the major row position, from a larger value to a smaller value.”) Overall, one of ordinary skill would have recognized that combining Badger with Bruguier and Jayaram would result in a system capable of generating a more varied initial set of input gauges/test patterns, leading to a more complete design space and ultimately enabling more accurately chosen sets of gauges to be selected and thereby more accurate calibrations.
Claim 17. Bruguier teaches wherein the instructions are further configured to cause the computer system to “[Col 9 line 8-12] “…a larger set of test patterns (containing K no. of test patterns, where K>K′) ….”) by use of user defined gauges ([Col 11 line 30-33] “The number of columns in the information matrix varies according to user-selected settings, and the particular computational lithography model used.” [Examiner’s note: the information matrix is used to define the gauges]) to determine the first subset of gauges. ([Col 9 line 8-12] “A general aspect of the invention is to select an optimal subset of test patterns (containing K′ no. of test patterns) in step 304 from a larger set of test patterns (containing K no. of test patterns, where K>K′) as described in more detail below”)
Jayaram makes obvious filter data by the use of user defined input ([Page 9 Par 2-3] “The following example shows the use of Union in two SELECT statements with a WHERE clause and ORDER BY clause. The following example is based on the rule 1,2 and 3” [Page 9 Bottom 2 Figures] The code blocks below obtain and filter data from the larger datasets)
PNG
media_image2.png
772
726
media_image2.png
Greyscale
Claim 22. Bruguier teaches wherein the instructions are further configured to cause the computer system to, ([Col 15 line 40-50] “FIG. 9 is an exemplary block diagram that illustrates a computer system 100 which can assist in embodying and/or implementing the pattern selection method disclosed herein. Computer system 100 includes a bus 102 or other communication mechanism for communicating information, and one or more processor(s) 104 (and 105) coupled with bus 102 for processing information. Computer system 100 also includes a main memory 106, such as a random access memory (RAM) or other dynamic storage device, coupled to bus 102 for storing information and instructions to be executed by processor 104”) ([Col 17 line 38-42] “FIG. 10 schematically depicts an exemplary lithographic projection apparatus whose performance could be simulated and/or optimized utilizing the computational lithography models that are calibrated using the test pattern selection process of present invention.”)
Jayaram makes obvious responsive to a determination that no duplicate gauges exist, select the merged subset of data. ([Page 3 Par 1] “The Union operator combines the results of two or more queries into a distinct single result set that includes all the rows that belong to all queries in the Union. In this operation, it combines two more queries and removes the duplicates.” [Page 9 Bottom 2 Figures] The code blocks merge data from the two subsets of the larger datasets, remove the duplicates, and select the results)
Claim 23. Bruguier teaches wherein the instructions are further configured to cause the computer system to calibrate the process model using the first subset of gauges. ([Col 17 line 38-42] “FIG. 10 schematically depicts an exemplary lithographic projection apparatus whose performance could be simulated and/or optimized utilizing the computational lithography models that are calibrated using the test pattern selection process of present invention.”)
Claim 35. Bruguier teaches a method comprising: obtaining a finite set of input gauges having one or more properties associated with a patterning process to physically form structures, the finite set having at least 100,000 input gauges and each input gauge corresponding to a pattern physically printed or configured for physical printing on a semiconductor substrate by the patterning process to form structures; selecting, by a hardware computer, a first subset of a plurality of gauges from the set of input gauges based on a first property parameter of the one or more properties; ([Col 6 line 8-23] “Embodiments of the present invention describes a method of selecting a subset of test patterns from an initial larger set of test patterns for calibrating a computational lithography model, the method comprising: generating an information matrix for the initial larger set of test patterns, wherein the terms of the information matrix comprise one or more identified model parameters that represent a lithographic process response; and, executing a selection algorithm using terms of the information matrix to select the subset of test patterns that effectively determines values of the identified model parameters that contribute significantly in the lithographic process response, wherein the subset of test patterns characteristically represents the initial larger set of test patterns. The selection algorithm explores coverage relationships existing in the initial larger set of test patterns. “[Col 9 line 8-12] “A general aspect of the invention is to select an optimal subset of test patterns (containing K′ no. of test patterns) in step 304 from a larger set of test patterns (containing K no. of test patterns, where K>K′) as described in more detail below” [Col 10 line 44- 49] “While step 402 is known in the art, the unique characteristic of the present invention is that the behavior of the solver is entirely determined by the input values of the parameters that are fed to the solver. In other words, a gauge can be entirely described by its corresponding input values in a general way, regardless of its geometry.” [Col 15 line 40-50] “FIG. 9 is an exemplary block diagram that illustrates a computer system 100 which can assist in embodying and/or implementing the pattern selection method disclosed herein”) determining a second subset, different from the first subset, of a plurality of gauges from the set of input gauges based on a second property parameter, different from the first property parameter, of the one or more properties; ([Col 9 line 42-55] “the test patterns are selected such that a selected test pattern is very sensitive to one or more specific model parameters, i.e., small changes in the parameters should be able to induce observable changes in the wafer CD for the test pattern. The test patterns are further selected such that the effect of different model parameters can be clearly distinguished. Test patterns with similar sensitivity to the model parameters are identified, grouped and selected such that no unnecessary duplication of test patterns is retained in the selection results. By achieving the selection above, the smallest possible set of test patterns is identified that achieves high sensitivity to the individual parameters as well as clear distinction between the contributions from different model parameters.” [Examiner’s note: creating groups of test patterns based on which parameters each test pattern is most sensitive to (i.e. a first group with a particular sensitivity to parameter A, a second group with a particular sensitivity to parameter B, etc.) is equivalent to creating different subsets each based on different parameters] [Col 10 line 44- 49] “While step 402 is known in the art, the unique characteristic of the present invention is that the behavior of the solver is entirely determined by the input values of the parameters that are fed to the solver. In other words, a gauge can be entirely described by its corresponding input values in a general way, regardless of its geometry.” [Examiner’s note: several input values are used, i.e. a first property parameter, second properties parameter, etc.]) ([Col 6 line 8-12] “… a subset of test patterns …”) and of the second subset of gauges([Col 9 line 42-55] “the test patterns are selected such that a selected test pattern is very sensitive to one or more specific model parameters, i.e., small changes in the parameters should be able to induce observable changes in the wafer CD for the test pattern. The test patterns are further selected such that the effect of different model parameters can be clearly distinguished. Test patterns with similar sensitivity to the model parameters are identified, grouped and selected such that no unnecessary duplication of test patterns is retained in the selection results. By achieving the selection above, the smallest possible set of test patterns is identified that achieves high sensitivity to the individual parameters as well as clear distinction between the contributions from different model parameters.” [Col 6 line 8-12] “… a subset of test patterns …”) to be a ([Col 6 line 8-12] “… a subset of test patterns …”) ([Col 6 line 8-12] “… a subset of test patterns …”) ([Col 6 line 8-12] “… a subset of test patterns …”) ([Col 6 line 8-12] “… a subset of test patterns …”) ([Col 6 line 8-12] “… a subset of test patterns …”) ([Col 6 line 8-12] “… a subset of test patterns …”) ([Col 6 line 8-12] “… a subset of test patterns …”) and applying at least the gauges by use of values from the third subset of gauges to change or configure one or more parameters of a process computer model that simulates an aspect of the patterning process([Col 17 line 38-42] “FIG. 10 schematically depicts an exemplary lithographic projection apparatus whose performance could be simulated and/or optimized utilizing the computational lithography models that are calibrated using the test pattern selection process of present invention.”) or by use of the reduced number of gauges to be measured in the third subset to produce a sampling plan for a metrology tool to make fewer measurements corresponding to the third subset of gauges than measuring all of the finite set of input gauges.
Bruguier does not explicitly teach the finite set having at least 100,000 input gauges; merging the data of the first subset and of the second subset to be a merged subset; determining if the merged subset of data include duplicate data; selecting a third subset of data from the merged subset such that the third subset does not include the duplicate data; the third subset of data
Jayaram makes obvious ([Page 3 Par 1] “The Union operator combines the results of two or more queries into a distinct single result set that includes all the rows that belong to all queries in the Union. In this operation, it combines two more queries and removes the duplicates.”)
PNG
media_image1.png
548
789
media_image1.png
Greyscale
Jayaram is analogous art because it is within the field of data selection. It would have been obvious to one of ordinary skill in the art to combine it with Bruguier before the effective filing date. As noted by Bruguier, the removal of duplicate test patterns is known to improve the model calibration process ([Col 9 line 46-55] “The test patterns are further selected such that the effect of different model parameters can be clearly distinguished. Test patterns with similar sensitivity to the model parameters are identified, grouped and selected such that no unnecessary duplication of test patterns is retained in the selection results. By achieving the selection above, the smallest possible set of test patterns is identified that achieves high sensitivity to the individual parameters as well as clear distinction between the contributions from different model parameters”) and that, generally, processing a smaller set of gauges without losing a significant amount of data (as by removing redundancies) results in a more efficient calibration process ([Col 10 line 4-17] “execute a selection algorithm to select the subset of K′ test patterns from the initial larger set of K patterns, where (K>K′), in a way so that the characteristics of all K gauges are sufficiently covered by the K′ test patterns … After having selected the subset of K′ patterns, the calibration of the lithographic model can proceed using conventional methods, except with fewer gauges (step 306), resulting in a computationally efficient calibration.”) Although Bruguier uses a mathematic method to distill the set of gauges into a simplified form, it is still plausible that duplicate gauges and their representations would remain in the reduced subset. To this end, Jayaram describes the UNION operator of SQL, which combines two datasets into a third dataset while removing the duplicates ([Page 3 Par 1] “The Union operator combines the results of two or more queries into a distinct single result set that includes all the rows that belong to all queries in the Union. In this operation, it combines two more queries and removes the duplicates.”) Overall, one of ordinary skill in the art would have recognized that combining Jayaram with Bruguier would result in a system that needs to process fewer gauges to perform calibration, therefore making the overall calibration process more efficient.
The combination of Bruguier and Jarayam does not explicitly teach the finite set having at least 100,000 input gauges
Badger makes obvious the finite set having at least 100,000 input gauges ([Par 27] “] The method for determining photomask inspection capabilities of the present invention includes first a method for generating test patterns, and a systematic method for arranging them on a mask data set. The method described will permit the generation of numerous, even millions, of different patterns in one mask” [Par 39] “As shown in FIG. 6, sub-arrays 26, each of about 1 mm .times.1 mm in dimension, for example, are arranged in a 9.times.10 array to form a main array 24. A plurality of main arrays 24, each of a size of about 9 mm .times.10 mm, are laid out in a 3.times.3 arrangement 22 which may contain about one million of the test patterns or shapes. Photomask 20 of the type used in lithographic production may have transferred thereon identical or different main arrays 22, in any other number, more or less, as desired.”)
Badger is analogous art because it is within the field of lithographic test pattern processing. It would have been obvious to one of ordinary skill in the art to combine Badger with Bruguier and Jayaram before the effective filing date. One of ordinary skill in the art would have been motivated to make this combination in order to generate test patterns in a way that more completely explores the possible design space, ultimately making any processing based on these patterns more accurate to reality and complete. Badger nodes how previous methods of generating test patterns suffer from too much uniformity, limiting the amount of design space that is explorable by systems processing the test patterns ([Par 7] “One way to identify inspection limits is to design test patterns, vary the dimensions of these test patterns, build a mask containing the patterns and then determine empirically which of these shapes passes mask inspection, as described in the U.S. Pat. Nos. 6,482,557 and 6,721,695. However, most of the patterns used in this approach are varied in an overly simple way across a pattern set. For example, the test patterns or shapes vary in one or two critical dimensions. This is due in part to the difficulties associated with mask inspection. When shapes that do not pass mask inspection are encountered, the tool flags the location. If too many of these flagged locations are present, the mask inspection is not completed. A large number of flagged inspection stops can be difficult to analyze efficiently, rendering the mask effectively uninspectable.”) To this end, Badger presents a method that introduces much deeper variety into the generated set of test patterns ([Par 29] “The flow chart of FIG. 1 provides an overview of the preferred method of the present invention. Test patterns or shapes are generated 10 that would include ordered variations of one or more shape variable, and the patterns are then formed and arranged on a mask 12 so as provide an opportunity to test variables or conditions of interest. The layout of the individual test patterns or shapes is important in practicing the method of the invention. The test patterns or shapes are systematically grouped by rule type with patterns of varying complexity, where complexity is measured by the number of variables that are modified within the patterns. [Par 37] “Four or more variables may also be tested, for example, by combining shape 30b with shape 30c. In this case of four variables, the variables would be nub width a, nub right side length b, nub left side length c, and nub distance to corner d. While the variations in nub dimensions a, b and c would be set out in the manner described in connection with basic arrays 28b. 28b' and 28b'', the dimension of the fourth variable d would be varied in the major row position, from a larger value to a smaller value.”) Overall, one of ordinary skill would have recognized that combining Badger with Bruguier and Jayaram would result in a system capable of generating a more varied initial set of input gauges/test patterns, leading to a more complete design space and ultimately enabling more accurately chosen sets of gauges to be selected and thereby more accurate calibrations.
Claims 18-20, 24, and 36 are rejected under 35 U.S.C. 103 as being unpatentable over Bruguier (US 9588439 B1) in view of SQL Union overview, usage and examples (hereinafter Jayaram) in further view of Badger (US 20070174012 A1) as well as Cao (NL 1036750 A1)
Claim 18. Cao teaches wherein the one or more properties comprises one or more selected from: a value of critical dimension of a substrate, ([Page 2 Par 3] “The design rule limitations could be referred to as "critical dimensions" (CD). A critical dimension of a circuit can be defined as the smallest width of a line or hole or the smallest space between two lines or two holes.”) a curvature associated with the pattern, and/or an intensity used in the patterning process.
Cao is analogous art because it is within the field of lithographic process calibration. It would have been obvious to one of ordinary skill in the art to combine Cao with Bruguier, Jayaram, and Badger before the effective filing date. One of ordinary skill in the art would have been motivated to make this combination in order to more accurately produce designs by further tuning the lithographic process. While Bruguier focuses on optimization of test pattern selection, Cao notes that another significant source of inaccuracy when tuning a lithography system is operation of the mask writing unit itself. ([Page 3 Par 2-3] “However, as each mask writer unit, equally identical model types, exhibit for example different proximity effects when generating a mask, the actual mask which is generated often differs from mask writer unit to mask writer unit. For example, different optical proximity effects (OPEs) associated with given optical mask writer units can introduce significant CD variations through pitch. As such, it is not possible to simply utilize any mask writer unit to generate a given mask, as the resulting mask can vary considerably. Thus, if it is desirable to utilize a different mask writer unit to form a given mask, the engineers must optimize or tune the new mask writer unit, so that the resulting mask formed by the mask writer unit satisfies the design requirements. Currently, this is typically accomplished by a trial and error process, which, as noted above, is both expensive and time consuming. As such, there is a need for a method for tuning or optimizing a given mask writer unit that allows the mask writer unit to produce a mask within a specified error tolerance relative to a previously tuned mask writer unit such that both mask writers are effectively capable of producing the same mask. In other words, there is a need for a method for optimizing the performance of multiple mask writers with respect to a given target mask that does not require a trial and error optimization process and which allows all mask writer units to produce masks within a predefined error tolerance.”) To this end, Cao presents a comprehensive system for tuning mask writers to produce consistent masks for lithographic processes and minimize variation([Page 3 Par 4] “… the present invention relates to a method for tuning mask writer units so as to allow different mask writer units to produce the same mask without requiring a substantial trial and error process to be performed to optimize the tunable parameters of each individual mask writer unit.” [Page 3 Par 7] “The model-based mask writer tuning process discussed follows provides numerous advantages about prior art methods. Most importantly, the present invention provides a systematic and cost effective method for the optimization of mask writing performance and mask proximity effect "MPE" matching between different mask writer units, which are utilized to generate the same mask. As a result, the present invention readily allows performance matching between different mask writers or the same model as well as for performance matching between different model mask writers.”) Overall, one of ordinary skill in the art would have recognized that combining Bruguier, Jayaram, and Badger with Cao would result in an overall lithographic manufacturing system that is produces more accurate, consistent results even across multiple physical production units.
Claim 19. Cao teaches wherein the first property parameter includes a model error, the model error being a difference between a reference contour and a simulated contour generated from a simulation using a process model of the patterning process. ([Page 7 Par 3] “Mask writer 504 may include a real device that in turn includes scanner 506 and wafer contour 508 and / or a virtual device that in turn includes model 510 and simulation contour 512. Similarly, turning target 502 may include a real device that in turn includes scanner 514 and wafer contour 516 and / or a virtual device that in turn comprises model 518 and simulation contour 520. As detailed above, the tuning amount is the difference between the performance (i.e., model) or the mask writer to be tuned and the performance (i.e., model) or the tuning target. In the model-based tuning process disclosed, the model (i.e., virtual mask writer) provides the link between the tuning target and the mask writer to be tuned.”)
Cao is analogous art because it is within the field of lithographic process calibration. It would have been obvious to one of ordinary skill in the art to combine Cao with Bruguier, Jayaram, and Badger before the effective filing date. One of ordinary skill in the art would have been motivated to make this combination in order to more accurately produce designs by further tuning the lithographic process. While Bruguier focuses on optimization of test pattern selection, Cao notes that another significant source of inaccuracy when tuning a lithography system is operation of the mask writing unit itself. ([Page 3 Par 2-3] “However, as each mask writer unit, equally identical model types, exhibit for example different proximity effects when generating a mask, the actual mask which is generated often differs from mask writer unit to mask writer unit. For example, different optical proximity effects (OPEs) associated with given optical mask writer units can introduce significant CD variations through pitch. As such, it is not possible to simply utilize any mask writer unit to generate a given mask, as the resulting mask can vary considerably. Thus, if it is desirable to utilize a different mask writer unit to form a given mask, the engineers must optimize or tune the new mask writer unit, so that the resulting mask formed by the mask writer unit satisfies the design requirements. Currently, this is typically accomplished by a trial and error process, which, as noted above, is both expensive and time consuming. As such, there is a need for a method for tuning or optimizing a given mask writer unit that allows the mask writer unit to produce a mask within a specified error tolerance relative to a previously tuned mask writer unit such that both mask writers are effectively capable of producing the same mask. In other words, there is a need for a method for optimizing the performance of multiple mask writers with respect to a given target mask that does not require a trial and error optimization process and which allows all mask writer units to produce masks within a predefined error tolerance.”) To this end, Cao presents a comprehensive system for tuning mask writers to produce consistent masks for lithographic processes and minimize variation([Page 3 Par 4] “… the present invention relates to a method for tuning mask writer units so as to allow different mask writer units to produce the same mask without requiring a substantial trial and error process to be performed to optimize the tunable parameters of each individual mask writer unit.” [Page 3 Par 7] “The model-based mask writer tuning process discussed follows provides numerous advantages about prior art methods. Most importantly, the present invention provides a systematic and cost effective method for the optimization of mask writing performance and mask proximity effect "MPE" matching between different mask writer units, which are utilized to generate the same mask. As a result, the present invention readily allows performance matching between different mask writers or the same model as well as for performance matching between different model mask writers.”) Overall, one of ordinary skill in the art would have recognized that combining Bruguier, Jayaram, and Badger with Cao would result in an overall lithographic manufacturing system that is produces more accurate, consistent results even across multiple physical production units.
Claim 20. Cao teaches wherein the reference contour is a measured contour from a scanning electron microscope. ([Page 10 Par 3] “ Finally, it is noted that the mask measurements necessary for the sensitivity-based tuning process discussed (as well as those mask measurements necessary for the model tuning of the first edition) can be performed utilizing, for example, CD-SEM ( e.g., CD measurements, and extracted mask contours from SEM pictures);”) [Examiner’s note: SEM stands for scanning electron microscope])
Claim 24. Bruguier teaches wherein the instructions are further configured to cause the computer system to ([Col 15 line 40-50] “FIG. 9 is an exemplary block diagram that illustrates a computer system 100 which can assist in embodying and/or implementing the pattern selection method disclosed herein. Computer system 100 includes a bus 102 or other communication mechanism for communicating information, and one or more processor(s) 104 (and 105) coupled with bus 102 for processing information. Computer system 100 also includes a main memory 106, such as a random access memory (RAM) or other dynamic storage device, coupled to bus 102 for storing information and instructions to be executed by processor 104”) ([Col 17 line 38-42] “FIG. 10 schematically depicts an exemplary lithographic projection apparatus whose performance could be simulated and/or optimized utilizing the computational lithography models that are calibrated using the test pattern selection process of present invention.”)
The combination of Bruguier, Jayaram, and Badger does not explicitly teach determine a process condition
Cao makes obvious determine a process condition ([Page 6 Par 6] “It is noted that the tunable parameters may include, but are not limited to, parameters associated with back-scatter correction, pattern dependent etch loading, fogging affects, average exposure, focus and current density. [Examiner’s note: the specification defines process condition as [Par 171] “The process condition comprises one or more process parameters, wherein the process parameter is at least one of: dose, focus, or intensity.”] )
Cao is analogous art because it is within the field of lithographic process calibration. It would have been obvious to one of ordinary skill in the art to combine Cao with Bruguier, Jayaram, and Badger before the effective filing date. One of ordinary skill in the art would have been motivated to make this combination in order to more accurately produce designs by further tuning the lithographic process. While Bruguier focuses on optimization of test pattern selection, Cao notes that another significant source of inaccuracy when tuning a lithography system is operation of the mask writing unit itself. ([Page 3 Par 2-3] “However, as each mask writer unit, equally identical model types, exhibit for example different proximity effects when generating a mask, the actual mask which is generated often differs from mask writer unit to mask writer unit. For example, different optical proximity effects (OPEs) associated with given optical mask writer units can introduce significant CD variations through pitch. As such, it is not possible to simply utilize any mask writer unit to generate a given mask, as the resulting mask can vary considerably. Thus, if it is desirable to utilize a different mask writer unit to form a given mask, the engineers must optimize or tune the new mask writer unit, so that the resulting mask formed by the mask writer unit satisfies the design requirements. Currently, this is typically accomplished by a trial and error process, which, as noted above, is both expensive and time consuming. As such, there is a need for a method for tuning or optimizing a given mask writer unit that allows the mask writer unit to produce a mask within a specified error tolerance relative to a previously tuned mask writer unit such that both mask writers are effectively capable of producing the same mask. In other words, there is a need for a method for optimizing the performance of multiple mask writers with respect to a given target mask that does not require a trial and error optimization process and which allows all mask writer units to produce masks within a predefined error tolerance.”) To this end, Cao presents a comprehensive system for tuning mask writers to produce consistent masks for lithographic processes and minimize variation([Page 3 Par 4] “… the present invention relates to a method for tuning mask writer units so as to allow different mask writer units to produce the same mask without requiring a substantial trial and error process to be performed to optimize the tunable parameters of each individual mask writer unit.” [Page 3 Par 7] “The model-based mask writer tuning process discussed follows provides numerous advantages about prior art methods. Most importantly, the present invention provides a systematic and cost effective method for the optimization of mask writing performance and mask proximity effect "MPE" matching between different mask writer units, which are utilized to generate the same mask. As a result, the present invention readily allows performance matching between different mask writers or the same model as well as for performance matching between different model mask writers.”) Overall, one of ordinary skill in the art would have recognized that combining Bruguier, Jayaram, and Badger with Cao would result in an overall lithographic manufacturing system that is produces more accurate, consistent results even across multiple physical production units.
Claim 36. Cao teaches wherein the first property parameter includes a model error, the model error being a difference between a reference contour and a simulated contour generated from a simulation using a process model of the patterning process. ([Page 7 Par 3] “Mask writer 504 may include a real device that in turn includes scanner 506 and wafer contour 508 and / or a virtual device that in turn includes model 510 and simulation contour 512. Similarly, turning target 502 may include a real device that in turn includes scanner 514 and wafer contour 516 and / or a virtual device that in turn comprises model 518 and simulation contour 520. As detailed above, the tuning amount is the difference between the performance (i.e., model) or the mask writer to be tuned and the performance (i.e., model) or the tuning target. In the model-based tuning process disclosed, the model (i.e., virtual mask writer) provides the link between the tuning target and the mask writer to be tuned.”)
Cao is analogous art because it is within the field of lithographic process calibration. It would have been obvious to one of ordinary skill in the art to combine Cao with Bruguier, Jayaram, and Badger before the effective filing date. One of ordinary skill in the art would have been motivated to make this combination in order to more accurately produce designs by further tuning the lithographic process. While Bruguier focuses on optimization of test pattern selection, Cao notes that another significant source of inaccuracy when tuning a lithography system is operation of the mask writing unit itself. ([Page 3 Par 2-3] “However, as each mask writer unit, equally identical model types, exhibit for example different proximity effects when generating a mask, the actual mask which is generated often differs from mask writer unit to mask writer unit. For example, different optical proximity effects (OPEs) associated with given optical mask writer units can introduce significant CD variations through pitch. As such, it is not possible to simply utilize any mask writer unit to generate a given mask, as the resulting mask can vary considerably. Thus, if it is desirable to utilize a different mask writer unit to form a given mask, the engineers must optimize or tune the new mask writer unit, so that the resulting mask formed by the mask writer unit satisfies the design requirements. Currently, this is typically accomplished by a trial and error process, which, as noted above, is both expensive and time consuming. As such, there is a need for a method for tuning or optimizing a given mask writer unit that allows the mask writer unit to produce a mask within a specified error tolerance relative to a previously tuned mask writer unit such that both mask writers are effectively capable of producing the same mask. In other words, there is a need for a method for optimizing the performance of multiple mask writers with respect to a given target mask that does not require a trial and error optimization process and which allows all mask writer units to produce masks within a predefined error tolerance.”) To this end, Cao presents a comprehensive system for tuning mask writers to produce consistent masks for lithographic processes and minimize variation([Page 3 Par 4] “… the present invention relates to a method for tuning mask writer units so as to allow different mask writer units to produce the same mask without requiring a substantial trial and error process to be performed to optimize the tunable parameters of each individual mask writer unit.” [Page 3 Par 7] “The model-based mask writer tuning process discussed follows provides numerous advantages about prior art methods. Most importantly, the present invention provides a systematic and cost effective method for the optimization of mask writing performance and mask proximity effect "MPE" matching between different mask writer units, which are utilized to generate the same mask. As a result, the present invention readily allows performance matching between different mask writers or the same model as well as for performance matching between different model mask writers.”) Overall, one of ordinary skill in the art would have recognized that combining Bruguier, Jayaram, and Badger with Cao would result in an overall lithographic manufacturing system that is produces more accurate, consistent results even across multiple physical production units.
Claims 25, 31-32, 34 are rejected under 35 U.S.C. 103 as being unpatentable over Bruguier (US 9588439 B1) in view of Cao (NL 1036750 A1) in further view of Zou (US 20100099032 A1) as well as Badger (US 20070174012 A1)
Claim 25. Bruguier makes obvious A computer program product comprising a non-transitory computer readable medium having instructions therein, the instructions, when executed by a computer system, configured to cause the computer system to at least: ([Col 15 line 40-50] “FIG. 9 is an exemplary block diagram that illustrates a computer system 100 which can assist in embodying and/or implementing the pattern selection method disclosed herein. Computer system 100 includes a bus 102 or other communication mechanism for communicating information, and one or more processor(s) 104 (and 105) coupled with bus 102 for processing information. Computer system 100 also includes a main memory 106, such as a random access memory (RAM) or other dynamic storage device, coupled to bus 102 for storing information and instructions to be executed by processor 104”) obtain a finite set of initial gauges having one or more properties associated with a patterning process to physically form structures, ([Col 9 line 8-12] “A general aspect of the invention is to select an optimal subset of test patterns (containing K′ no. of test patterns) in step 304 from a larger set of test patterns (containing K no. of test patterns, where K>K′) as described in more detail below” [Col 10 line 44- 49] “While step 402 is known in the art, the unique characteristic of the present invention is that the behavior of the solver is entirely determined by the input values of the parameters that are fed to the solver. In other words, a gauge can be entirely described by its corresponding input values in a general way, regardless of its geometry.”) ([Col 9 line 8-12] “A general aspect of the invention is to select an optimal subset of test patterns (containing K′ no. of test patterns) in step 304 from a larger set of test patterns (containing K no. of test patterns, where K>K′) as described in more detail below” ([Col 9 line 8-12] “A general aspect of the invention is to select an optimal subset of test patterns (containing K′ no. of test patterns) in step 304 from a larger set of test patterns (containing K no. of test patterns, where K>K′) as described in more detail below” [Col 8 line 55- Col 9 line 2] “Pattern selection does not change the physical and chemical effects contained in the model, but should preferably help to adequately excite these effects such that the degrees to which they manifest themselves in a specific lithography process can be efficiently identified based on the wafer measurements for the selected test patterns. The insight here is that the method of determining what is an effective subset of patterns may depend on many factors, such as: whether the model is based on first-principles or is an empirical model, whether the goal is to select the fewest patterns to calibrate the model to make the process most efficient computationally, whether a predefined predictive accuracy goal is set, whether a customer has provided a preferred set of test patterns that must be included in any subset, etc. “) and([Col 17 line 38-42] “FIG. 10 schematically depicts an exemplary lithographic projection apparatus whose performance could be simulated and/or optimized utilizing the computational lithography models that are calibrated using the test pattern selection process of present invention.”) or by use of the reduced number of gauges to be measured in the subset of gauges to produce a sampling plan for a metrology tool to make fewer measurements corresponding to the subset of gauges than measurement of all of the finite set of initial input gauges.
Bruguier does not explicitly teach the finite set having at least 100,000 initial gauges, calibrate, via an optimization algorithm, a plurality of models, each model of the plurality of models being associated with a model error value after calibration; determine one or more candidate models from the plurality of models based on a comparison of the model error value associated with each model with respect to a model error value of a particular model in the plurality of models; select a subset of gauges based on information generated from the one or more candidate models;
Cao makes obvious calibrate, via an optimization algorithm, ([Page 5 Par 4] “As is known, during the calibration process, which is an iterative process, the non-tunable parameters are fixed and the tunable parameters are adjusted until the mask generated by the model (i.e., the simulated mask result) matches the actual mask result produced by the reference mask writer. Thus, the model parameters MP R are adjusted (i.e., calibrated) such that the mask results produced by the model equal the actual mask data associated with the reference mask writer MD R within some predefined error criteria or the best match possible.”) a plurality of models, ([Page 6 par 4] “Once the model MP R is tuned for each of the mask writers, creating "n" models MPRI ..... MPRn (where n is the number of mask writers), in Step 50, each of the mask writers is tuned from the nominal parameter values that were utilized to generate the initial wafer data MD_1 .... MD n utilizing the parameter values of the reference model MP_R and the adjusted model parameters MPRi… The resulting parameters, Pi, are then utilized to tune the corresponding mask writer (i),”) each model of the plurality of models being associated with a model error value after calibration; ([Page 6 Par 1] “It is noted that the foregoing model has achieved accuracy of 3σ <3 nm (mask scale) for 1D patterns on several 65 nm and 45 nm masks manufactured with different processes. On 2D patterns, model error 3σ is typically about 10 nm for both calibration and prediction.”) determine one or more candidate models from the plurality of models based on a comparison of the model error value associated with each model with respect to a model error value of a particular model in the plurality of models; ([Page 5 Par 4] “As is known, during the calibration process, which is an iterative process, the non-tunable parameters are fixed and the tunable parameters are adjusted until the mask generated by the model (i.e., the simulated mask result) matches the actual mask result produced by the reference mask writer. Thus, the model parameters MP R are adjusted (i.e., calibrated) such that the mask results produced by the model equal the actual mask data associated with the reference mask writer MD R within some predefined error criteria or the best match possible … It is further noted that any one of the mask writers is tuned and associated mask data, MD 1 ... MD-n, may be utilized to calibrate the model (i.e., may be utilized as the referenced mask writer)” [Page 6 par 4] “Once the model MP R is tuned for each of the mask writers, creating "n" models MPRI ..... MPRn (where n is the number of mask writers), in Step 50, each of the mask writers is tuned from the nominal parameter values that were utilized to generate the initial wafer data MD_1 .... MD n utilizing the parameter values of the reference model MP_R and the adjusted model parameters MPRi… The resulting parameters, Pi, are then utilized to tune the corresponding mask writer (i),”) ([Page 5 Par 4] “As is known, during the calibration process, which is an iterative process, the non-tunable parameters are fixed and the tunable parameters are adjusted until the mask generated by the model (i.e., the simulated mask result) matches the actual mask result produced by the reference mask writer. Thus, the model parameters MP R are adjusted (i.e., calibrated) such that the mask results produced by the model equal the actual mask data associated with the reference mask writer MD R within some predefined error criteria or the best match possible … It is further noted that any one of the mask writers is tuned and associated mask data, MD 1 ... MD-n, may be utilized to calibrate the model (i.e., may be utilized as the referenced mask writer)”)
Cao is analogous art because it is within the field of lithographic process calibration. It would have been obvious to one of ordinary skill in the art to combine Cao with Bruguier before the effective filing date. One of ordinary skill in the art would have been motivated to make this combination in order to more accurately produce designs by further tuning the lithographic process. While Bruguier focuses on optimization of test pattern selection, Cao notes that another significant source of inaccuracy when tuning a lithography system is operation of the mask writing unit itself. ([Page 3 Par 2-3] “However, as each mask writer unit, equally identical model types, exhibit for example different proximity effects when generating a mask, the actual mask which is generated often differs from mask writer unit to mask writer unit. For example, different optical proximity effects (OPEs) associated with given optical mask writer units can introduce significant CD variations through pitch. As such, it is not possible to simply utilize any mask writer unit to generate a given mask, as the resulting mask can vary considerably. Thus, if it is desirable to utilize a different mask writer unit to form a given mask, the engineers must optimize or tune the new mask writer unit, so that the resulting mask formed by the mask writer unit satisfies the design requirements. Currently, this is typically accomplished by a trial and error process, which, as noted above, is both expensive and time consuming. As such, there is a need for a method for tuning or optimizing a given mask writer unit that allows the mask writer unit to produce a mask within a specified error tolerance relative to a previously tuned mask writer unit such that both mask writers are effectively capable of producing the same mask. In other words, there is a need for a method for optimizing the performance of multiple mask writers with respect to a given target mask that does not require a trial and error optimization process and which allows all mask writer units to produce masks within a predefined error tolerance.”) To this end, Cao presents a comprehensive system for tuning mask writers to produce consistent masks for lithographic processes and minimize variation([Page 3 Par 4] “… the present invention relates to a method for tuning mask writer units so as to allow different mask writer units to produce the same mask without requiring a substantial trial and error process to be performed to optimize the tunable parameters of each individual mask writer unit.” [Page 3 Par 7] “The model-based mask writer tuning process discussed follows provides numerous advantages about prior art methods. Most importantly, the present invention provides a systematic and cost effective method for the optimization of mask writing performance and mask proximity effect "MPE" matching between different mask writer units, which are utilized to generate the same mask. As a result, the present invention readily allows performance matching between different mask writers or the same model as well as for performance matching between different model mask writers.”) Overall, one of ordinary skill in the art would have recognized that combining Bruguier with Cao would result in an overall lithographic manufacturing system that is produces more accurate, consistent results even across multiple physical production units.
The combination of Bruguier and Cao does not explicitly teach the finite set having at least 100,000 initial gauges, selection based on information generated from a model
Zou makes obvious selection based on information generated from a model ([Par 31] “The process begins at step 501 by simulating a lithographic production of a full chip layout. The layout is analyzed lithographically at step 503 to identify printability failure points, or hot spots. The hot spots represent the areas where the presence of mask assist features could improve process latitude and or other manufacturability metrics. These areas are analyzed for layout configuration patterns, and the patterns are classified (step 505). Pattern classification is accomplished by breaking down layout configurations into clusters of layout shapes based on geometrical similarity criteria. Unique patterns are extracted, and duplicate patterns are eliminated. These unique layout patterns are selected as representative layout configurations for the full chip layout at step 507”)
Zou is analogous art because it is within the field of semiconductor manufacturing through the use of lithography. One of ordinary skill in the art would have been motivated to combine it with Bruguier and Cao before the effective filing date. One of ordinary skill in the art would have been motivated to make this combination in order to produce a more accurate lithography system. As semiconductor components continue to get smaller and smaller, traditional methods of forming nanoscale integrated circuit structures begin to fail as they bump up against the very laws of physics themselves, as Zou puts succinctly ([Par 3] “ There is a continuing objective to increase the density with which various integrated circuit structures are arranged. As technology nodes decrease, the critical dimension of the features to be printed on the silicon wafer is reduced. As the size of the features becomes smaller than the wavelength of light, distortions occur in the printed patterns.”) To overcome these limitations, Zou introduces mask assist features; additional features beyond the printed features themselves that aid in the formation of the desired structures. ([Par 3] “ To reduce these distortions, mask assist features (AF) are added to the mask between the features to be printed. AF are not printed on the semiconductor wafer, but help to balance the optical density of the feature pattern. ([Par 4] “ AF are special geometrical figures, or polygons, which are added (or sometimes subtracted) to design layouts in lithographic processes for manufacturing integrated circuits and other related fields such as hard-disk heads, etc. AF are also referred to as sub-resolution assist features (SRAF), scatter bars (SB), and in some cases printable assist features (PRAF),”) One of ordinary skill in the art would recognize that the accuracy increase afforded by integrating the features of Zou with those of Bruguier and Cao would produce a far improved gauge selection system, allowing even finer tuning and even greater levels of calibration.
The combination of Bruguier, Cao, and Zou does not explicitly teach the finite set having at least 100,000 initial gauges
Badger makes obvious the finite set having at least 100,000 initial gauges ([Par 27] “] The method for determining photomask inspection capabilities of the present invention includes first a method for generating test patterns, and a systematic method for arranging them on a mask data set. The method described will permit the generation of numerous, even millions, of different patterns in one mask” [Par 39] “As shown in FIG. 6, sub-arrays 26, each of about 1 mm .times.1 mm in dimension, for example, are arranged in a 9.times.10 array to form a main array 24. A plurality of main arrays 24, each of a size of about 9 mm .times.10 mm, are laid out in a 3.times.3 arrangement 22 which may contain about one million of the test patterns or shapes. Photomask 20 of the type used in lithographic production may have transferred thereon identical or different main arrays 22, in any other number, more or less, as desired.”)
Badger is analogous art because it is within the field of lithographic test pattern processing. It would have been obvious to one of ordinary skill in the art to combine Badger with Bruguier, Cao, and Zou before the effective filing date. One of ordinary skill in the art would have been motivated to make this combination in order to generate test patterns in a way that more completely explores the possible design space, ultimately making any processing based on these patterns more accurate to reality and complete. Badger nodes how previous methods of generating test patterns suffer from too much uniformity, limiting the amount of design space that is explorable by systems processing the test patterns ([Par 7] “One way to identify inspection limits is to design test patterns, vary the dimensions of these test patterns, build a mask containing the patterns and then determine empirically which of these shapes passes mask inspection, as described in the U.S. Pat. Nos. 6,482,557 and 6,721,695. However, most of the patterns used in this approach are varied in an overly simple way across a pattern set. For example, the test patterns or shapes vary in one or two critical dimensions. This is due in part to the difficulties associated with mask inspection. When shapes that do not pass mask inspection are encountered, the tool flags the location. If too many of these flagged locations are present, the mask inspection is not completed. A large number of flagged inspection stops can be difficult to analyze efficiently, rendering the mask effectively uninspectable.”) To this end, Badger presents a method that introduces much deeper variety into the generated set of test patterns ([Par 29] “The flow chart of FIG. 1 provides an overview of the preferred method of the present invention. Test patterns or shapes are generated 10 that would include ordered variations of one or more shape variable, and the patterns are then formed and arranged on a mask 12 so as provide an opportunity to test variables or conditions of interest. The layout of the individual test patterns or shapes is important in practicing the method of the invention. The test patterns or shapes are systematically grouped by rule type with patterns of varying complexity, where complexity is measured by the number of variables that are modified within the patterns. [Par 37] “Four or more variables may also be tested, for example, by combining shape 30b with shape 30c. In this case of four variables, the variables would be nub width a, nub right side length b, nub left side length c, and nub distance to corner d. While the variations in nub dimensions a, b and c would be set out in the manner described in connection with basic arrays 28b. 28b' and 28b'', the dimension of the fourth variable d would be varied in the major row position, from a larger value to a smaller value.”) Overall, one of ordinary skill would have recognized that combining Badger with Bruguier, Cao, and Zou would result in a system capable of generating a more varied initial set of input gauges/test patterns, leading to a more complete design space and ultimately enabling more accurately chosen sets of gauges to be selected and thereby more accurate calibrations.
Claim 31. Cao teaches wherein the one or more properties comprise at least one selected from: a value of critical dimension of a substrate, ([Page 2 Par 3] “The design rule limitations could be referred to as "critical dimensions" (CD). A critical dimension of a circuit can be defined as the smallest width of a line or hole or the smallest space between two lines or two holes.”) a curvature associated with the pattern, and/or an intensity used in the patterning process.
Claim 32. Cao teaches wherein the model error value associated with each model corresponds to a difference between a reference contour and a simulated contour generated from a simulation using a process model of the patterning process, ([Page 7 Par 3] “Mask writer 504 may include a real device that in turn includes scanner 506 and wafer contour 508 and / or a virtual device that in turn includes model 510 and simulation contour 512. Similarly, turning target 502 may include a real device that in turn includes scanner 514 and wafer contour 516 and / or a virtual device that in turn comprises model 518 and simulation contour 520. As detailed above, the tuning amount is the difference between the performance (i.e., model) or the mask writer to be tuned and the performance (i.e., model) or the tuning target. In the model-based tuning process disclosed, the model (i.e., virtual mask writer) provides the link between the tuning target and the mask writer to be tuned.”) the reference contour being a measured contour from an image capture device. ([Page 5 Par 3] “It is noted the measurement of the mask data may be made, for example, by performing various CD measurements or measuring portions or the entire contour of the imaged feature which can be performed utilizing a SEM.” [Examiner’s note: SEM stands for Scanning Electron Microscope])
Claim 34. Bruguier teaches wherein the instructions are further configured to cause the computer system to [Col 15 line 40-50] “FIG. 9 is an exemplary block diagram that illustrates a computer system 100 which can assist in embodying and/or implementing the pattern selection method disclosed herein. Computer system 100 includes a bus 102 or other communication mechanism for communicating information, and one or more processor(s) 104 (and 105) coupled with bus 102 for processing information. Computer system 100 also includes a main memory 106, such as a random access memory (RAM) or other dynamic storage device, coupled to bus 102 for storing information and instructions to be executed by processor 104”) ([Col 17 line 38-42] “FIG. 10 schematically depicts an exemplary lithographic projection apparatus whose performance could be simulated and/or optimized utilizing the computational lithography models that are calibrated using the test pattern selection process of present invention.”)
Cao makes obvious determine a process condition ([Page 6 Par 6] “It is noted that the tunable parameters may include, but are not limited to, parameters associated with back-scatter correction, pattern dependent etch loading, fogging affects, average exposure, focus and current density. [Examiner’s note: the specification defines process condition as [Par 171] “The process condition comprises one or more process parameters, wherein the process parameter is at least one of: dose, focus, or intensity.”] )
Claims 26-27 are rejected under 35 U.S.C. 103 as being unpatentable over Bruguier (US 9588439 B1) in view of Cao (NL 1036750 A1) in further view of Zou (US 20100099032 A1) as well as Badger (US 20070174012 A1) and SQL Union overview, usage and examples (hereinafter Jayaram)
Claim 26. Bruguier teaches wherein the instructions configured to cause the computer system ([Col 15 line 40-50] “FIG. 9 is an exemplary block diagram that illustrates a computer system 100 which can assist in embodying and/or implementing the pattern selection method disclosed herein. Computer system 100 includes a bus 102 or other communication mechanism for communicating information, and one or more processor(s) 104 (and 105) coupled with bus 102 for processing information. Computer system 100 also includes a main memory 106, such as a random access memory (RAM) or other dynamic storage device, coupled to bus 102 for storing information and instructions to be executed by processor 104”) to obtain the initial gauges are further configured to cause the computer system to: determine a first subset of gauges from the initial gauges based on a first property of the one or more properties, ([Col 6 line 8-23] “Embodiments of the present invention describes a method of selecting a subset of test patterns from an initial larger set of test patterns for calibrating a computational lithography model, the method comprising: generating an information matrix for the initial larger set of test patterns, wherein the terms of the information matrix comprise one or more identified model parameters that represent a lithographic process response; and, executing a selection algorithm using terms of the information matrix to select the subset of test patterns that effectively determines values of the identified model parameters that contribute significantly in the lithographic process response, wherein the subset of test patterns characteristically represents the initial larger set of test patterns. The selection algorithm explores coverage relationships existing in the initial larger set of test patterns. “[Col 9 line 8-12] “A general aspect of the invention is to select an optimal subset of test patterns (containing K′ no. of test patterns) in step 304 from a larger set of test patterns (containing K no. of test patterns, where K>K′) as described in more detail below” [Col 10 line 44- 49] “While step 402 is known in the art, the unique characteristic of the present invention is that the behavior of the solver is entirely determined by the input values of the parameters that are fed to the solver. In other words, a gauge can be entirely described by its corresponding input values in a general way, regardless of its geometry.”)([Col 6 line 8-12] “Embodiments of the present invention describes a method of selecting a subset of test patterns from an initial larger set of test patterns for calibrating a computational lithography model,”) based on a second property of the one or more properties; ([Col 10 line 44- 49] “While step 402 is known in the art, the unique characteristic of the present invention is that the behavior of the solver is entirely determined by the input values of the parameters that are fed to the solver. In other words, a gauge can be entirely described by its corresponding input values in a general way, regardless of its geometry.” [Examiner’s note: several input values are used, i.e. a first property parameter, second properties parameter, etc.]) ([Col 6 line 8-12] “… a subset of test patterns …”) and the ([Col 6 line 8-12] “… a subset of test patterns …”) to be a ([Col 6 line 8-12] “… a subset of test patterns …”) ([Col 6 line 8-12] “… a subset of test patterns …”) ([Col 6 line 8-12] “… a subset of test patterns …”) and ([Col 6 line 8-12] “… a subset of test patterns …”) ([Col 6 line 13-16] “wherein the terms of the information matrix comprise one or more identified model parameters that represent a lithographic process response”) ([Col 6 line 8-12] “… a subset of test patterns …”)
Cao makes obvious the first property being a weight and/or a model error; ([Page 6 Par 1] “It is noted that the foregoing model has achieved accuracy of 3σ <3 nm (mask scale) for 1D patterns on several 65 nm and 45 nm masks manufactured with different processes. On 2D patterns, model error 3σ is typically about 10 nm for both calibration and prediction.” [Page 5 Par 4] “As is known, during the calibration process, which is an iterative process, the non-tunable parameters are fixed and the tunable parameters are adjusted until the mask generated by the model (i.e., the simulated mask result) matches the actual mask result produced by the reference mask writer. Thus, the model parameters MP R are adjusted (i.e., calibrated) such that the mask results produced by the model equal the actual mask data associated with the reference mask writer MD R within some predefined error criteria or the best match possible”)
The combination of Bruguier, Cao, Zou, and Badger does not explicitly teach determine a second subset of data; merge the first subset of data and the second subset of data to be a merged subset of data; determine if the merged subset of data include duplicate data; and selecting a third subset of the merged subset of data based on one or more properties such that the third subset does not include the duplicate data.
Jayaram makes obvious determine a second subset of data; merge the first subset of data and the second subset of data to be a merged subset of data; determine if the merged subset of data include duplicate data; and selecting a third subset of the merged subset of data based on one or more properties such that the third subset does not include the duplicate data. ([Page 3 Par 1] “The Union operator combines the results of two or more queries into a distinct single result set that includes all the rows that belong to all queries in the Union. In this operation, it combines two more queries and removes the duplicates.”)
PNG
media_image1.png
548
789
media_image1.png
Greyscale
Jayaram is analogous art because it is within the field of data selection. It would have been obvious to one of ordinary skill in the art to combine it with Bruguier, Cao, Zou, and Badger before the effective filing date. As noted by Bruguier, the removal of duplicate test patterns is known to improve the model calibration process ([Col 9 line 46-55] “The test patterns are further selected such that the effect of different model parameters can be clearly distinguished. Test patterns with similar sensitivity to the model parameters are identified, grouped and selected such that no unnecessary duplication of test patterns is retained in the selection results. By achieving the selection above, the smallest possible set of test patterns is identified that achieves high sensitivity to the individual parameters as well as clear distinction between the contributions from different model parameters”) and that, generally, processing a smaller set of gauges without losing a significant amount of data (as by removing redundancies) results in a more efficient calibration process ([Col 10 line 4-17] “execute a selection algorithm to select the subset of K′ test patterns from the initial larger set of K patterns, where (K>K′), in a way so that the characteristics of all K gauges are sufficiently covered by the K′ test patterns … After having selected the subset of K′ patterns, the calibration of the lithographic model can proceed using conventional methods, except with fewer gauges (step 306), resulting in a computationally efficient calibration.”) To this end, Jayaram describes the UNION operator of SQL, which combines two datasets into a third dataset while removing the duplicates ([Page 3 Par 1] “The Union operator combines the results of two or more queries into a distinct single result set that includes all the rows that belong to all queries in the Union. In this operation, it combines two more queries and removes the duplicates.”) Overall, one of ordinary skill in the art would have recognized that combining Jayaram with Bruguier, Cao, Zou, and Badger would result in a system that needs to process fewer gauges to perform calibration, therefore making the overall calibration process more efficient.
Claim 27. Bruguier teaches wherein the instructions are further configured to cause the computer system to([Col 15 line 40-50] “FIG. 9 is an exemplary block diagram that illustrates a computer system 100 which can assist in embodying and/or implementing the pattern selection method disclosed herein. Computer system 100 includes a bus 102 or other communication mechanism for communicating information, and one or more processor(s) 104 (and 105) coupled with bus 102 for processing information. Computer system 100 also includes a main memory 106, such as a random access memory (RAM) or other dynamic storage device, coupled to bus 102 for storing information and instructions to be executed by processor 104”)([Col 6 line 8-23] “Embodiments of the present invention describes a method of selecting a subset of test patterns from an initial larger set of test patterns for calibrating a computational lithography model”) by use of user defined gauges ([Col 11 line 30-33] “The number of columns in the information matrix varies according to user-selected settings, and the particular computational lithography model used.” [Examiner’s note: the information matrix is used to define the gauges])to determine the first subset of gauges ([Col 9 line 8-12] “A general aspect of the invention is to select an optimal subset of test patterns (containing K′ no. of test patterns) in step 304 from a larger set of test patterns (containing K no. of test patterns, where K>K′) as described in more detail below”)and the ([Col 6 line 8-12] “… a subset of test patterns …”)
Jayaram makes obvious filtering data by the use of user defined input to determine the second subset of data ([Page 9 Par 2-3] “The following example shows the use of Union in two SELECT statements with a WHERE clause and ORDER BY clause. The following example is based on the rule 1,2 and 3” [Page 9 Bottom 2 Figures] The code blocks below obtain and filter data from the larger datasets. The second SELECT statement is equivalent to determining the second subset)
PNG
media_image2.png
772
726
media_image2.png
Greyscale
Claims 28-30, 33 are rejected under 35 U.S.C. 103 as being unpatentable over Bruguier (US 9588439 B1) in view of Cao (NL 1036750 A1) in further view of Zou (US 20100099032 A1) as well as Badger (US 20070174012 A1) and Ke (US 20150079700 A1)
Claim 28. Bruguier teaches wherein the instructions are further configured to cause the computer system to ([Col 15 line 40-50] “FIG. 9 is an exemplary block diagram that illustrates a computer system 100 which can assist in embodying and/or implementing the pattern selection method disclosed herein. Computer system 100 includes a bus 102 or other communication mechanism for communicating information, and one or more processor(s) 104 (and 105) coupled with bus 102 for processing information. Computer system 100 also includes a main memory 106, such as a random access memory (RAM) or other dynamic storage device, coupled to bus 102 for storing information and instructions to be executed by processor 104”)
Cao makes obvious ([Page 5 Par 4] “As is known, during the calibration process, which is an iterative process, the non-tunable parameters are fixed and the tunable parameters are adjusted until the mask generated by the model (i.e., the simulated mask result) matches the actual mask result produced by the reference mask writer. Thus, the model parameters MP R are adjusted (i.e., calibrated) such that the mask results produced by the model equal the actual mask data associated with the reference mask writer MD R within some predefined error criteria or the best match possible … It is further noted that any one of the mask writers is tuned and associated mask data, MD 1 ... MD-n, may be utilized to calibrate the model (i.e., may be utilized as the referenced mask writer)” [Page 6 par 4] “Once the model MP R is tuned for each of the mask writers, creating "n" models MPRI ..... MPRn (where n is the number of mask writers)…”)
The combination of Bruguier, Cao, Zou, and Badger does not explicitly teach determine a similarity metric between at least two models.
Ke makes obvious determine a similarity metric between at least two models. ([Par 23] “In step 314, the first model correction map and the second model correction map are used to generate a similarity index. In step 316, a determination is made as to whether the first and second model correction maps are sufficiently or insufficiently similar”)
Ke is analogous art because it is within the field of semiconductor manufacturing with a particular focus on lithography. It would have been obvious to one of ordinary skill in the art to combine it with Bruguier, Cao, Zou, and Badger before the effective filing date. One of ordinary skill in the art would have been motivated to make this combination in order to simplify the semiconductor manufacturing process. As pointed out by Ke, ([Par 4] “ The pursuit of smaller feature size has required a number of technological changes, including changes in the control of fabrication processes. In a semiconductor fabrication facility, often referred to as a "fab," monitoring the results of process steps has become even more critical. Misalignment, lithography defects, and tool drift can result in a process generating unsatisfactory results even after a period of time with satisfactory results”) The large quantity of process steps utilizing a multitude of different tools introduces a chance of failure that increases exponentially with every step. By the time the fabrication process reaches the lithography step, it may have already failed, making any adjustments or calibrations to enhance the lithographic process moot. To overcome this, Ke presents a system that monitors production at every step and detects when deviations or abnormalities are introduced ([Par 22] “ When the similarity index is outside specifications, e.g. below the pre-determined threshold, an alert may be issued or further wafer production may be placed on hold to allow modification to the set of process parameters in step 310. The alert may be provided by a dedicated light, a sound, and/or text appearing in a process control system interface. The set of process parameters depends on the tool that is used in a particular process step being monitored.”) so appropriate corrective measures can be taken before time and resources are wasted continuing work on an already faulty wafer. One of ordinary skill in the art would have recognized that by combining Ke with Bruguier, Cao, Zou, and Badger they could create a process that ensures that the wafer presented to the lithographic system for which gauges are selected has been sufficiently processed beforehand.
Claim 29. Cao teaches ([Page 5 Par 4] “As is known, during the calibration process, which is an iterative process, the non-tunable parameters are fixed and the tunable parameters are adjusted until the mask generated by the model (i.e., the simulated mask result) matches the actual mask result produced by the reference mask writer. Thus, the model parameters MP R are adjusted (i.e., calibrated) such that the mask results produced by the model equal the actual mask data associated with the reference mask writer MD R within some predefined error criteria or the best match possible … It is further noted that any one of the mask writers is tuned and associated mask data, MD 1 ... MD-n, may be utilized to calibrate the model (i.e., may be utilized as the referenced mask writer)” [Page 6 par 4] “Once the model MP R is tuned for each of the mask writers, creating "n" models MPRI ..... MPRn (where n is the number of mask writers)…”)
Ke makes obvious wherein the similarity metric is a cosine similarity metric being a cosine of two vectors, ([Par 20] “The similarity index provides a correlation of the similarities between two sets of vectors in different wafers. In some embodiments, the correlation function used to obtain the similarity index is a cosine similarity function.”) each vector being representative of a given model of the models. ([Fig. 3] Shows two different sets of metrology data, which each contain vector data, being associated with the models [Par 23] “In step 314, the first model correction map and the second model correction map are used to generate a similarity index. In step 316, a determination is made as to whether the first and second model correction maps are sufficiently or insufficiently similar”)
Claim 30. Bruguier teaches wherein the instructions are further configured to cause the computer system to ([Col 15 line 40-50] “FIG. 9 is an exemplary block diagram that illustrates a computer system 100 which can assist in embodying and/or implementing the pattern selection method disclosed herein. Computer system 100 includes a bus 102 or other communication mechanism for communicating information, and one or more processor(s) 104 (and 105) coupled with bus 102 for processing information. Computer system 100 also includes a main memory 106, such as a random access memory (RAM) or other dynamic storage device, coupled to bus 102 for storing information and instructions to be executed by processor 104”)
Cao makes obvious wherein the instructions are further configured to cause the computer system to ([Page 5 Par 4] “As is known, during the calibration process, which is an iterative process, the non-tunable parameters are fixed and the tunable parameters are adjusted until the mask generated by the model (i.e., the simulated mask result) matches the actual mask result produced by the reference mask writer. Thus, the model parameters MP R are adjusted (i.e., calibrated) such that the mask results produced by the model equal the actual mask data associated with the reference mask writer MD R within some predefined error criteria or the best match possible … It is further noted that any one of the mask writers is tuned and associated mask data, MD 1 ... MD-n, may be utilized to calibrate the model (i.e., may be utilized as the referenced mask writer)” [Page 6 par 4] “Once the model MP R is tuned for each of the mask writers, creating "n" models MPRI ..... MPRn (where n is the number of mask writers)…”)([Page 5 Par 4] “As is known, during the calibration process, which is an iterative process, the non-tunable parameters are fixed and the tunable parameters are adjusted until the mask generated by the model (i.e., the simulated mask result) matches the actual mask result produced by the reference mask writer. Thus, the model parameters MP R are adjusted (i.e., calibrated) such that the mask results produced by the model equal the actual mask data associated with the reference mask writer MD R within some predefined error criteria or the best match possible … It is further noted that any one of the mask writers is tuned and associated mask data, MD 1 ... MD-n, may be utilized to calibrate the model (i.e., may be utilized as the referenced mask writer)” [Page 6 par 4] “Once the model MP R is tuned for each of the mask writers, creating "n" models MPRI ..... MPRn (where n is the number of mask writers)…”) ([Page 5 Par 4] “As is known, during the calibration process, which is an iterative process, the non-tunable parameters are fixed and the tunable parameters are adjusted until the mask generated by the model (i.e., the simulated mask result) matches the actual mask result produced by the reference mask writer. Thus, the model parameters MP R are adjusted (i.e., calibrated) such that the mask results produced by the model equal the actual mask data associated with the reference mask writer MD R within some predefined error criteria or the best match possible”)
Ke makes obvious select, based on the similarity metric, a diverse model, wherein the diverse model has a value of the similarity metric substantially different from a value of the similarity metric of another model ([Par 23] “In step 314, the first model correction map and the second model correction map are used to generate a similarity index. In step 316, a determination is made as to whether the first and second model correction maps are sufficiently or insufficiently similar. This involves comparing the similarity index to a threshold value. When the similarity index resulting from the correlation of the first and second model correction maps is within specifications, e.g. greater than or equal to a pre-determined threshold, the computer-generated wafer model is updated in step 318. The model may be updated with either the first or second model correction map…” [Examiner’s note: one of ordinary skill in the art would have recognized that being able to detect when a model is insufficiently similar to another and knowing a model with a least model error value would allow finding a model that is insufficiently similar to the model with the least model error value])
Claim 33. Bruguier makes obvious wherein the selection of the gauges is based on ([Col 9 line 8-12] “A general aspect of the invention is to select an optimal subset of test patterns (containing K′ no. of test patterns) in step 304 from a larger set of test patterns (containing K no. of test patterns, where K>K′) as described in more detail below” [Col 10 line 2-4] “execute a selection algorithm to select the subset of K′ test patterns from the initial larger set of K patterns, where (K>K′),”)
The combination of Bruguier, Cao, Zou, and Badger does not explicitly teach at least one selected from: a mean value of the model error, a standard deviation value of the model error, and/or a peak-to-peak value of the model
Ke makes obvious at least one selected from: a mean value of the model error, ([Fig. 7] [Par 31] “As matrix 700 is illustrated, the similarity indices are also depicted with the x-direction and y-direction mean error measurements.”) a standard deviation value of the model error, and/or a peak-to-peak value of the model [Examiner’s note: this claim is written in an alternative form; therefore the unmapped limitations are not given patentable weight.]
Ke is analogous art because it is within the field of semiconductor manufacturing with a particular focus on lithography. It would have been obvious to one of ordinary skill in the art to combine it with Bruguier, Cao, Zou, and Badger before the effective filing date. One of ordinary skill in the art would have been motivated to make this combination in order to simplify the semiconductor manufacturing process. As pointed out by Ke, ([Par 4] “ The pursuit of smaller feature size has required a number of technological changes, including changes in the control of fabrication processes. In a semiconductor fabrication facility, often referred to as a "fab," monitoring the results of process steps has become even more critical. Misalignment, lithography defects, and tool drift can result in a process generating unsatisfactory results even after a period of time with satisfactory results”) The large quantity of process steps utilizing a multitude of different tools introduces a chance of failure that increases exponentially with every step. By the time the fabrication process reaches the lithography step, it may have already failed, making any adjustments or calibrations to enhance the lithographic process moot. To overcome this, Ke presents a system that monitors production at every step and detects when deviations or abnormalities are introduced ([Par 22] “ When the similarity index is outside specifications, e.g. below the pre-determined threshold, an alert may be issued or further wafer production may be placed on hold to allow modification to the set of process parameters in step 310. The alert may be provided by a dedicated light, a sound, and/or text appearing in a process control system interface. The set of process parameters depends on the tool that is used in a particular process step being monitored.”) so appropriate corrective measures can be taken before time and resources are wasted continuing work on an already faulty wafer. One of ordinary skill in the art would have recognized that by combining Ke with Bruguier, Cao, Zou, and Badger they could create a process that ensures that the wafer presented to the lithographic system for which gauges are selected has been sufficiently processed beforehand.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Michael P Mirabito whose telephone number is (703)756-1494. The examiner can normally be reached M-F 10:30 am - 6:30 pm.
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, Emerson Puente can be reached at (571) 272-3652. 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.
/M.P.M./ Examiner, Art Unit 2187
/EMERSON C PUENTE/ Supervisory Patent Examiner, Art Unit 2187
Prosecution Insights