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 .
Status of the Application
2. Claim 1-20 have been examined in this application. This communication is the first action on the merits.
Drawings
3. The drawings filed on 12/15/23 are acceptable for examination proceedings.
Claim Rejections - 35 USC § 103
4. 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.
5. Claim 1-4, 7-11, 14-18 are rejected under 35 U.S.C. 103 as being unpatentable over Banna (Pub: 2020/0110390) in view of Nakata (Pub: 2019/0236463).
6. Regarding claim 1, Banna teaches a method, comprising: providing, to a trained machine learning model configured to determine a recommended adjustment to an equipment constant of a substrate manufacturing system (e.g., FIG. 2 and FIG. 3 show the two basic capabilities of the machine-learning based model. When metrology data 202 is used to generate final spatial model(s) 112, the machine-learning based model can predict spatial dimensions of interest 215 based on various process recipes and control knob information 211. On the other hand, when reference spatial measurements (sometimes called “golden profiles”) 302 are used as input, an inverse spatial model 312 can recommend recipe 315 for a given process and given equipment (chosen from a database of processes and equipment) when the process/equipment information 313 is fed to the inverse spatial model 312) (Para. [0055], also refer to Para. [0074]), first input data indicative of a state of the substrate manufacturing system (e.g., The machine-learning method used by the machine-learning engine 108 can be based on neural network, deep learning or any other known techniques used for regression analysis (e.g., linear, partial least squares, Gaussian, polynomials, convolution neural networks for regression, regression trees and others). In addition to metrology data, the machine-learning engine 108 also receives information 111 about various recipes and knobs, as well as information 113 about the process and the equipment. The machine-learning engine 108 then generates intermediate spatial model 109 for each measurement on the wafer. Each measurement can have data about one or more dimensions of interest) (Para. [0052], Fig. 1B- element 113 as a process and equipment info);
providing, to the trained machine learning model as second input data, an indication of a performed adjustment to the equipment constant (e.g., The machine-learning method used by the machine-learning engine 108 can be based on neural network, deep learning or any other known techniques used for regression analysis (e.g., linear, partial least squares, Gaussian, polynomials, convolution neural networks for regression, regression trees and others). In addition to metrology data, the machine-learning engine 108 also receives information 111 about various recipes and knobs, as well as information 113 about the process and the equipment. The machine-learning engine 108 then generates intermediate spatial model 109 for each measurement on the wafer. Each measurement can have data about one or more dimensions of interest) (Para. [0052], Fig. 1B);
Banna does not specifically teach and retraining the trained machine learning model based on a difference between the recommended adjustment to the equipment constant and the performed adjustment to the equipment constant to generate a retrained machine learning model.
Nakata teaches and retraining the trained machine learning model based on a difference between the recommended adjustment to the equipment constant and the performed adjustment to the equipment constant to generate a retrained machine learning model (e.g., A data processing system according to an aspect of the present disclosure, including the computer: inputs, in the third trained model, the first output data corresponding to the first input data for the first trained model to obtain the second output data, the third trained model being acquired through training in which (i) the output data of the first trained model is used as training data, and (ii) the output data of the second trained model acquired by converting the first trained model is used as label data; obtains the first label data of the first input data; and retrains the first trained model using the first differential data corresponding to differences between the second output data and the first label data) (Para. [0038]).
Because Nakata is also directed to a data processing system for executing machine learning and a data processing method, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having teachings of Banna and Nakata before him/her, to modify the teachings of Banna to include the retraining the learned machine learning model teaching of Nakata in order to enhancing the performance of the trained model after having been converted even when the contents of the conversion tool for converting the trained model are unknown (Nakata: Para. [00005]).
7. Regarding claim 2, the combination of Banna and Nakata teaches the method of claim 1, wherein Banna further comprising training a machine learning model to generate the trained machine learning model, wherein training the machine learning model comprises: providing data indicative of a plurality of states of the substrate manufacturing system as training input (e.g., Aspects of the disclosure describe a method and a corresponding system for controlling chamber-to-chamber variability during manufacturing of a device on wafers. Specifically, a computer-implemented method is described, where, for each current chamber in a multi-chamber processing platform, a spatial model for a wafer is obtained. The spatial model is created by a first machine-learning engine based on a first set of metrology data on one or more dimensions of interest in the device. The spatial model can be global, covering physical behavior of the process, or could be chamber-specific, accounting for chamber variability. One or more parameters of the current chamber are obtained. Using the spatial model and the one or more parameters of the current chamber, spatial measurements of the one or more dimensions of interest in the device across the wafer are predicted) (Para. [0007]);
providing data indicative of a plurality of adjustments to the equipment constant corresponding to the plurality of states of the substrate manufacturing system as target output (e.g., During the research and development phase, the disclosed systems and methods provide for faster convergence to target process recipes using only a limited number of test wafers. During production ramp leading to high volume manufacturing (HVM), the disclosed systems and methods enables tighter control of the process window not only intra-wafer, but also between wafers in a single lot (wafer-to-wafer control), or between different lots of wafers (lot-to-lot control). The process control frequency and frequency of model adaptation may vary depending on whether it is wafer-to-wafer (higher frequency), lot-to-lot (medium frequency) or PM-to-PM (i.e. once at each periodic maintenance (PM))) (Para. [0031]);
and training the machine learning model based on the training input and the target output (e.g., FIG. 2 and FIG. 3 show the two basic capabilities of the machine-learning based model. When metrology data 202 is used to generate final spatial model(s) 112, the machine-learning based model can predict spatial dimensions of interest 215 based on various process recipes and control knob information 211. On the other hand, when reference spatial measurements (sometimes called “golden profiles”) 302 are used as input, an inverse spatial model 312 can recommend recipe 315 for a given process and given equipment (chosen from a database of processes and equipment) when the process/equipment information 313 is fed to the inverse spatial model 312. One or both of these two capabilities can be used during the model training and calibration phase as well as during run-time wafer-to-wafer variability control phase. Spatial measurement prediction is more useful during the calibration process, and recipe prediction is more useful during the wafer-to-wafer control phase (for example, maintaining and/or optimizing a process of record (POR) for the HVM phase), as discussed further below in the specification) (Para. [0055]).
8. Regarding claim 3, the combination of Banna and Nakata teaches the method of claim 2, wherein Banna further teaches the target output comprises output from a rule-based algorithm (e.g., The model uses machine-learning algorithms to combine all the data and extract meaningful relationships between metrology of dimensions of interest and various knobs that control the process) (Para. [0039]) or calculation based on the data indicative of the plurality of states of the substrate manufacturing system (e.g., In block 1070, the suggestion from block 1030 is received and a corrected recipe is calculated for the next wafer run. The recipe correction may be recommended to an advanced process control (APC) host) (Para. [0080], also refer to Para. [0074])
9. Regarding claim 4, the combination of Nakata and Banna teaches the method of claim 1, wherein Banna further comprising determining a penalty based on the performed adjustment and the recommended adjustment (e.g., The model's performance is evaluated by the evaluation module 110. The model's performance is optimized using a penalty function or cost function 105, such as root mean square error (rMSE) or any other suitable metric.) (Para. [0052], also refer to Para. [0046]-[0047]),
wherein [retraining] the trained machine learning model comprises updating one or more parameters of the trained machine learning model based on the penalty (e.g., The model's performance is evaluated by the evaluation module 110. The model's performance is optimized using a penalty function or cost function 105, such as root mean square error (rMSE) or any other suitable metric. The cost function is sometimes referred to as “objective function,” designed to allow optimization of one or more dimensions of interest. The cost function can be for each location on a wafer, or just one cost function for an entire wafer. Cost function can also be for each DoE condition. Optimization routines (including, but not limited to swarm optimization or swarm variants, are designed to minimize non-convex multi-minima hyper-surfaces. Error penalties or regularization terms may be added to the cost function to find higher probability solutions in high dimension non-convex multi-minima hyper-surfaces. Once the desired value of the cost function is obtained, the spatial model may be further validated using metrology data from another set of physical DoE wafers. Number of test and validation wafers may be in the range of tens or twenties, but may vary. Depending on how a cost function is chosen, the test and validation process can be repeated spatially for each data point across a wafer for which metrology was conducted. Alternatively, the spatial model can be optimized to achieve an average dimensional uniformity across the wafer. The final spatial model 112 can combine results from all the data points on the wafer for which metrology was conducted) (Para. [0052]).
10. Regarding claim 7, the combination of Nakata and Banna teaches the method of claim 1, wherein Banna further teaches the substrate manufacturing system comprises a semiconductor wafer processing tool (e.g., Aspects of the disclosure describe a method and a corresponding system for controlling chamber-to-chamber variability during manufacturing of a device on wafers. Specifically, a computer-implemented method is described, where, for each current chamber in a multi-chamber processing platform, a spatial model for a wafer is obtained) (Para. [0007]).
11. Regarding claim 8, Claim 8 recites a non-transitory machine-readable storage medium that implement the method of claim 1, with substantially the same limitations, respectively. Therefore the rejection applied to claim 1, also applies to claim 8 respectively.
Wherein Banna further teaches a non-transitory machine-readable storage medium, storing instructions which, when executed, cause a processing device to perform operations (e.g., The example computer system 1300 includes a processing device 1302, a main memory 1304 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous DRAM (SDRAM) etc.), a static memory 1306 (e.g., flash memory, static random access memory (SRAM), etc.), and a data storage device 1316, which communicate with each other via a bus 1308) (Para. [0086]-
[0087]).
12. Regarding claim 9 and 16, as to claim 9 and 16, applicant is directed to citation of claim 2 above.
13. Regarding claim 10 and 17, as to claim 10 and 17, applicant is directed to citation of claim 3 above.
14. Regarding claim 11, as to claim 11, applicant is directed to citation of claim 4 above.
15. Regarding claim 14, as to claim 14, applicant is directed to citation of claim 7 above.
16. Regarding claim 15, Claim 15 recites a system that implement the method of claim 1, with substantially the same limitations, respectively. Therefore the rejection applied to claim 1, also applies to claim 15 respectively.
Wherein Banna further teaches a system, comprising memory and a processing device coupled to the memory, wherein the processing device (e.g., The example computer system 1300 includes a processing device 1302, a main memory 1304 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous DRAM (SDRAM) etc.), a static memory 1306 (e.g., flash memory, static random access memory (SRAM), etc.), and a data storage device 1316, which communicate with each other via a bus 1308) (Para. [0086]-
[0087]).
17. Regarding claim 18, as to claim 18, applicant is directed to citation of claim 4 above.
18. Claim 5-6, 12-13, and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Banna in view of Nakata, and further in view of Sadeghi (Pub: 2022/0270901).
19. Regarding claim 5, the combination of Banna and Nakata teaches a method of claim 1 but does not specifically teach wherein the performed adjustment is associated with input to the substrate manufacturing system based on one or more of user input or feedback input to update the equipment constant.
Sadeghi teaches wherein the performed adjustment is associated with input to the substrate manufacturing system based on one or more of user input or feedback input to update the equipment constant (e.g., The systems and methods complement various sensors, computer vision algorithms, and feedback mechanisms typically used to determine health of the process modules and tools and to recommend measures to control the tool and improve throughput, yield, and on-wafer process quality. The systems and methods continue to learn from images being captured on an ongoing basis. The systems and methods automate performance of some of the preventive and corrective tasks that typically require human intervention. The systems and methods oversee tool operations across the fleet in a semiconductor fab and provide feedback for autonomous control of the tool) (Para. [0074], also refer to Para. [0307] for user interface that enables entry or programming of parameters and/or settings).
Because Sadeghi is also directed to substrate processing systems, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having teachings of Banna, Nakata and Sadeghi before him/her, to modify the combined teachings of Banna, and Nakata to include the user input or feedback input teaching of Sadeghi for closed-loop autonomous control of the tool and adjust processes running in the PMs (Sadeghi: Para. [0159]).
20. Regarding claim 6, the combination of Banna and Nakata teaches a method of claim 1 but does not specifically teach wherein the equipment constant comprises a parameter associated with ring height of a ring in a process chamber configured to perform substrate etching operations.
Sadeghi teaches wherein the equipment constant comprises a parameter associated with ring height of a ring in a process chamber configured to perform substrate etching operations (e.g., The system computer 2410 automatically adjusts the recipe used in a PM based on reduction in thickness of an incoming and outgoing edge coupling ring. The system computer 2410 determines an etch rate in a PM based on a detected change in inner diameter (i.e., change in edge erosion profile) or thickness of incoming/outgoing edge coupling rings. The system computer 2410 uses this information to automatically adjust a recipe in a PM for tool-to-tool PM matching or to adjust the height of the edge coupling ring above the chuck by adjusting the lift pins. These and other functions performed by the computer vision system 2400 based on the images captured by the cameras 2402 are now described in detail) (Para. [0172]).
Because Sadeghi is also directed to substrate processing systems, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having teachings of Banna, Nakata and Sadeghi before him/her, to modify the combined teachings of Banna, and Nakata to include the ring height of a ring in a process chamber teaching of Sadeghi for autonomous control of the tool for PM matching via edge coupling ring height measurement and control (Sadeghi: Para. [0197]).
21. Regarding claim 12, and 19, as to claim 12 and 19, applicant is directed to citation of claim 5 above.
22. Regarding claim 13 and 20, as to claim 13 and 20, applicant is directed to citation of claim 6 above.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JIGNESHKUMAR C PATEL whose telephone number is (571)270-0698. The examiner can normally be reached Monday - Friday, 7:00 AM - 5:00 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, Kenneth M. Lo can be reached at (571)272-9774. 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.
/JIGNESHKUMAR C PATEL/Primary Examiner, Art Unit 2116