Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
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.
35 U.S.C. 101 requires that a claimed invention must fall within one of the four eligible categories of invention (i.e., process, machine, manufacture, or composition of matter) and must not be directed to subject matter encompassing a judicially recognized exception as interpreted by the courts. MPEP 2106. The four eligible categories of invention include: (1) process which is an act, or a series of acts or steps, (2) machine which is an concrete thing, consisting of parts, or of certain devices and combination of devices, (3) manufacture which is an article produced from raw or prepared materials by giving to these materials new forms, qualities, properties, or combinations, whether by hand labor or by machinery, and (4) composition of matter which is all compositions of two or more substances and all composite articles, whether they be the results of chemical union, or of mechanical mixture, or whether they be gases, fluids, powders or solids. MPEP 2106(I).
Claims 1-14, are rejected under 35 U.S.C. 101 abstract idea, while the claims recite a series of steps or acts to be performed, as an example such as, “acquiring a plurality of shot images of a wafer after performing a first process on the wafer using a first layout; generating an overlap image obtained by overlap of the plurality of shot images; and performing machine learning on the proximity correction model by using an image of the first layout and the overlap image”.
Prong 1 analysis:
The steps do not amount to significantly more than the abstract idea. The recited steps could be implemented by the user or a human operator (mental activity), observing an image, recited steps “acquiring a plurality of shot images of a wafer after performing a first process on the wafer using a first layout”. Accordingly, the analysis under prong one of step 2A of the Subject Matter Eligibility Test does not result in a conclusion of eligibility (See flowchart MPEP 2106).
Prong 2 analysis:
The additional elements regarding claim 1, “performing machine learning on the proximity correction model by using an image”. could be implemented by the user further gathering, analyzing data and organizing by “mathematical manipulation or algorithm”, does not preempt all possible ways for correction based on the images of the object, or any specific way to implement calculating. The steps do not amount to significantly more than the abstract idea, they are recited at a high level of generality and are conventional, well known and routine. The claim as a whole is an abstract idea. Accordingly, the analysis under step 2B of the Subject Matter Eligibility Test does not result in a conclusion of eligibility (See flowchart MPEP 2106).
DETAILED ACTION
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1, 6-11, 17 and 19-20 are rejected under 35 U.S.C. 102(a) (1) as being anticipated by Cho et al (U.S. Pub. No: 2022/0292669 A1).
Regarding claim 1, Cho discloses a method for generating a proximity correction model which is performed by a computing system, the method comprising (see page 1, paragraphs, [0006-0007], there is provided a method of providing an EUV photomask. The method includes providing a design layout, performing an “optical proximity correction” (OPC) on the design layout, verifying the OPC, and manufacturing an EUV photomask responsive to determining that a result of the verifying of the OPC is correct, in which the verifying of the result of the OPC includes generating a contour histogram image, based on the design layout and a resist image, an aerial image, a slope map, a density map, and/or a photon map corresponding to the design layout, and a value of a pixel included in the contour histogram image indicates a probability that a contour of patterns included in the design layout is located in the pixel. According to some embodiments, there is provided a system configured to stochastically predict a defect caused by a “lithography process”. The system includes a stochastic prediction model configured to generate a first contour histogram image);
acquiring a plurality of shot images of a wafer after performing a first process on the wafer using a first layout; generating an overlap image obtained by overlap of the plurality of shot images (see abstract, a method of providing a stochastic prediction system. The method includes extracting contours of patterns corresponding to a first design layout from a plurality of scanning electron microscope (SEM) images (shot images), respectively, generating a first contour histogram image based on the contours, and training a stochastic prediction model by using the first contour histogram image as an output, and by using the first design layout and a first resist image, a first aerial image, a first slope map, a first density map, and/or a first photo map corresponding to the first design layout as inputs, in which the stochastic prediction model comprises a cycle generative adversarial network (GAN).
Also, page 2, paragraphs, [0018-0019] the SEM image may be an image of a photoresist pattern generated by after-development inspection or an image of a real circuit pattern generated by after-clean inspection. FIG. 2 shows an overlapping of different contours by merging a contour image extracted from an SEM image with respect to the same region of different wafers or different chips of the same wafer. Referring to FIG. 3, a “lithography process” will be described. FIG. 3 shows a process of manufacturing an integrated circuit through a lithography semiconductor process);
and performing machine learning on the proximity correction model by using an image of the first layout and the overlap image (see above, also page 5, paragraphs, [0065-0066] OPC may be performed. OPC may include rule-based OPC, model-based OPC, and/or machine learning-guided OPC. The rule-based OPC may be driven by a lookup table that is calculated in advance based on a width and an interval between features. The model-based OPC may be driven using a compact model to dynamically simulate a final pattern, and an optimal solution may be found by inducing bias (movement) of an edge divided into sections. The machine learning-guided OPC may be a method of directly obtaining a mask image on which OPC is performed from a target layout, by using a machine-learning algorithm, without performing iterative lithography simulation. When one segment of a design layout expressed as a parameter (e.g., a pattern density, an optical kernel signal, etc.) is input to a pre-trained artificial neural network, bias of the segment may be output. As the artificial neural network determines and applies bias for all segments, an OPC mask image may be generated. In some cases, model-based OPC may be performed successively from machine learning-guided OPC, thereby improving the speed of OPC without lowering the accuracy of OPC. Performing model-based OPC may include executing and/or simulating a process model by using an initial mask image as an input and generating a process image to be formed on a wafer. Herein, the process image may include, for example, the aerial image, the resist image, the etching image, etc. In an embodiment, the process model may include a “mask” transmission model coupled to an optical model).
Regarding claim 6, Cho discloses the method for generating the proximity correction model of claim 1, wherein the proximity correction model is an image-to-image model which is configured to receive an image of the first layout, and to output a predicted image of a pattern formed corresponding to a shape of the first layout after performing the first process (see claim 1, also, page 3, paragraph, [0033] light passing through an opening of the photomask PM may be diffracted. As the size of patterns of the photomask PM decreases, an optical proximity effect (OPE) may occur due to an influence between adjacent ones of the patterns. To compensate for an error caused by the diffraction and the OPE described above, optical proximity correction (OPC) may be adopted. For example, as shown in the right side of FIG. 3, to form the first pattern P11, a second pattern P12 to which OPC is applied may be formed on the photomask PM, and the second pattern P12 may have a shape that is different from that of the first pattern P11. The second pattern P12 may have a shape that is corrected by OPC. Also, page 5, paragraphs, [0071-0073] performing OPC may include adding sub-lithographic features referred to as serifs at a corner of a pattern or adding sub-lithographic assist features such as scattering bars, as well as transformation of a layout of the pattern. The serif, which is a rectangular feature at each corner of a pattern, may be used to “sharpen” corners of the pattern or compensate for a distortion factor caused by intersection of the pattern. A sub resolution assist feature (SRAF), which is an auxiliary feature introduced to solve an OPC deviation problem caused by a density difference of the pattern, may be formed to a size less than a resolution of exposure equipment and thus may not be transferred to a resist layer. Next, referring to FIGS. 6A and 7, in P130, OPC may be verified by the stochastic prediction model 10. Verification of OPC, performed by the stochastic prediction model 10, may include generating a contour histogram image, by the stochastic prediction model 10, based on at least any one of a design layout that is a target of OPC or a resist image, an aerial image, a slope map, a density map, or a photo map corresponding to the design layout).
Regarding claim 7, Cho discloses the method for generating the proximity correction model of claim 1, wherein the generating the overlap image comprises extracting contours of each of plurality of shot images, dithering the extracted contours of each of the plurality of shot images, and generating the overlap image by overlapping the dithered contours of each of the plurality of shot images (see claim 1, also page 1, paragraphs, [0005-0006] the inventive concepts also provide a method of providing a stochastic prediction system. The method of providing a system for stochastically predicting a contour includes “extracting contours” of patterns corresponding to a first design layout from a plurality of scanning electron microscope (SEM) images (shot images), respectively, generating a first contour histogram image based on the contours, and training a stochastic prediction model by using the first contour histogram image as an output, and by using the first design layout and a first resist image, a first aerial image, a first slope map, a first density map, and/or a first photo map corresponding to the first design layout as inputs, in which the stochastic prediction model includes a cycle generative adversarial network (GAN). According to some embodiments, there is provided a method of providing an EUV photomask. The method includes providing a design layout, performing an optical proximity correction (OPC) on the design layout, verifying the OPC, and manufacturing an EUV photomask responsive to determining that a result of the verifying of the OPC is correct, in which the verifying of the result of the OPC includes generating a contour histogram image, based on the design layout and a resist image, an aerial image, a slope map, a density map, and/or a photon map corresponding to the design layout, and a value of a pixel included in the contour histogram image indicates a probability that a contour of patterns included in the design layout is located in the pixel.
Also, page 1, paragraphs, [0017-0018] referring to FIGS. 1 and 2, in operation P10, a “contour may be extracted” from images obtained through a scanning electron microscope (SEM). An SEM image may be generated from SEM equipment or NGR equipment of NGR Inc. (formerly NanoGeometry® Research). In operation P10, a binary image indicating a contour of a measured layout may be generated. Herein, the SEM image may be an image of a photoresist pattern generated by after-development inspection or an image of a real circuit pattern generated by after-clean inspection. FIG. 2 shows an overlapping of different contours by merging a contour image extracted from an SEM image with respect to the same region of different wafers or different chips of the same wafer).
Regarding claim 9, Cho discloses the method for generating the proximity correction model of claim 1, wherein the performing machine learning comprises: performing the machine learning on the proximity correction model, by further using a representative image, representing the plurality of shot images, as a correct image (see claim 1, also page 4, paragraph, [0049] the stochastic prediction model 10 may include an artificial neural network, a decision tree, a support vector machine, a Bayesian network, and/or a genetic algorithm, etc. Hereinbelow, an embodiment where the stochastic prediction model 10 is an artificial neural network will be mainly described. This is merely for convenience of a description, and does not limit the scope of the inventive concepts. The artificial neural network may include, for example, a convolution neural network (CNN), a region with a CNN (R-CNN), a fast R-CNN, a region proposal network (RPN), a recurrent neural network (RNN), a stacking-based deep neural network (S-DNN), a state-space dynamic neural network (S-SDNN), a deconvolution network, a deep belief network (DBN), a restricted Boltzmann machine (RBM), a full convolutional network, a long short-term memory (LSTM) network, a classification network, etc. According to some embodiments, the stochastic prediction model 10 may be implemented by, for example, a neural processing unit (NPU), a graphic processing unit (GPU), etc.).
Regarding claim 10, Cho discloses the method for generating the proximity correction model of claim 9, wherein the representative image is an image representing an average contour of the plurality of shot images (see claim 1, also see abstract, the inventive concepts provide a method of providing a stochastic prediction system. The method includes extracting contours of patterns corresponding to a first design layout from a plurality of scanning electron microscope (SEM) images (shot image), respectively, generating a first contour histogram image based on the contours, and training a stochastic prediction model by using the first contour histogram image as an output, and by using the first design layout and a first resist image, a first aerial image, a first slope map, a first density map, and/or a first photo map corresponding to the first design layout as inputs, in which the stochastic prediction model comprises a cycle generative adversarial network (GAN). Also, page 5, paragraph, [0065-0066] next, in operation P120, OPC may be performed. OPC may include rule-based OPC, model-based OPC, and/or machine learning-guided OPC. The rule-based OPC may be driven by a lookup table that is calculated in advance based on a width and an interval between features. The model-based OPC may be driven using a compact model to dynamically simulate a final pattern, and an optimal solution may be found by inducing bias (movement) of an edge divided into sections. The machine learning-guided OPC may be a method of directly obtaining a mask image on which OPC is performed from a target layout, by using a machine-learning algorithm, without performing iterative lithography simulation. When one segment of a design layout expressed as a parameter (e.g., a pattern density, an optical kernel signal, etc.) is input to a pre-trained artificial neural network, bias of the segment may be output. As the artificial neural network determines and applies bias for all segments, an OPC mask image may be generated. In some cases, model-based OPC may be performed successively from machine learning-guided OPC, thereby improving the speed of OPC without lowering the accuracy of OPC. Performing model-based OPC may include executing and/or simulating a process model by using an initial mask image as an input and generating a process image to be formed on a wafer. Herein, the process image may include, for example, the aerial image, the resist image, the etching image, etc. In an embodiment, the process model may include a mask transmission model coupled to an optical model, which is additionally coupled to a resist model and/or an etching model. An output of the process model may be a process image considering a process change during a simulation process. The process image may be additionally used to determine a parameter (e.g., an edge placement error (EPE), a critical dimension, an overlay, a side lobe, etc.) of a patterning process by tracking a contour of a pattern in the process image).
Regarding claim 11, Cho discloses the method for generating the proximity correction model of claim 9, wherein the performing the machine learning on the proximity correction model, by further using the representative image representing the plurality of shot images comprises: updating parameters of the proximity correction model by performing an error back-propagation using loss data, the loss data representing a difference between an image output by the proximity correction model and the representative image (see claim 1, also page 7, paragraphs, [0090-0093] format conversion, i.e., fracturing, may mean a process of changing the MTO design data into a format for an electron beam exposure by fracturing the MTO design data for each region. Fracturing may include data manipulation, for example, scaling, data sizing, data rotation, pattern radiation, color inversion, etc. In conversion through fracturing, data regarding numerous systematic errors likely to occur in delivery from design data to an image on a wafer may be corrected. A data correction process for the systematic errors may be referred to as mask process correction (MPC), and may include line width adjustment called CD adjustment and a task of improving the accuracy of pattern placement. Thus, fracturing may contribute to improvement of the quality of a final mask and may be proactively performed for mask process correction. Herein, systematic errors may be caused by distortion occurring in an exposure process, a mask development and etching process, a wafer imaging process, etc. The MDP may include the MPC. The MPC may refer to a process of correcting an error occurring in the exposure process, i.e., a systematic error. Herein, the exposure process may be a concept including overall electron beam writing, development, etching, baking, etc. In addition, data processing may be performed before the exposure process. The data processing refers to a preprocessing process regarding a sort of mask data, and may include a grammar check for the mask data, exposure time prediction, etc. After the MDP, a mask wafer may be exposed based on the mask data. Herein, exposure may mean, for example, electron beam writing. The electron beam writing may be performed in a gray writing manner using, for example, a multi-beam mask writer (MBMW). The electron beam writing may be performed using a variable shape beam (VSB) exposure. After mask data preparation, a process of converting mask data into pixel data before exposure may be performed. The pixel data may be data directly used for actual exposure, and may include data regarding a shape of an exposure target and data regarding a dose assigned to each data. Herein, the data regarding the shape may be bit-map data converted from shape data, which is vector data, through rasterization, etc.).
Regarding claim 19, Cho discloses the proximity correction method of the manufacturing process of claim 17, wherein the first layout is a layout of a photoresist pattern, the first process is an etching process, the pattern formed after performing the first process is an etching pattern, the target image is an ACI (After Cleaning Inspection) target image, and the performing the proximity correction process comprises adjusting the layout of the photoresist pattern based on a difference between the ACI target image and a predicted image of the etching pattern (see claim 1, also page 3, paragraphs, [0035-0036] etching may be performed in a third structure 13, such that a part of the oxide layer that is not protected by the photoresist may be etched. Etching may include wet (or liquid) etching and dry (or plasma) etching. Etching may remove a part of a topmost layer that is not protected by the photoresist. After completion of etching, the photoresist may be removed by a cleaning process, such that as shown in FIG. 3, a fourth pattern P14 may be formed on the oxide layer. According to some embodiments, the wafer may be heated to remove a residual solvent through a cleaning process. As shown in FIG. 3, as sub processes are performed, the shape of a pattern may be transformed, which may be modelled through pattern transformation. For example, the first pattern P11 corresponding to the ideal pattern may be transformed into the second pattern P12 through application of OPC, and the second pattern P12 may be transformed into the third pattern P13 in an after-development inspection (ADI) state through irradiation and development. The third pattern P13 may be transformed into a fourth pattern P14 in an after clean inspection (ACI) state through etching and cleaning. As a result, the fourth pattern P14 may have a shape that is different from the ideal pattern, i.e., the first pattern P11, but it may be important to determine a shape of the second pattern P12 such that the fourth pattern P14 may have a shape that is more or most similar to the first pattern).
With regard to claims 8, 17 and 20 the arguments analogous to those presented above for claims 1, 6, 7, 9, 10, 11 and 19, are respectively applicable to claims 8, 17 and 20.
Allowable Subject Matter
Claims 2-5, 12-16 and 18 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Contact Information
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Seyed Azarian whose telephone number is (571) 272-7443. The examiner can normally be reached on Monday through Thursday from 6:00 a.m. to 7:30 p.m.
If attempts to reach the examiner by telephone are unsuccessful, the examiner's supervisor, Andrew Moyer, can be reached at (571) 272-9523. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application information Retrieval (PAIR) system. Status information for published application may be obtained from either Private PAIR or Public PAIR.
Status information about the PAIR system, see http:// pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
/SEYED H AZARIAN/Primary Examiner, Art Unit 2667
February 7, 2026