A machine learning approach to model the impact of line edge roughness on gate-all-around nanowire FETs while reducing the carbon footprint

Antonio García-Loureiro; Natalia Seoane; Julián G. Fernández; Enrique Comesaña; Juan C. Pichel

doi:10.1371/journal.pone.0288964

Peer Review History

Original SubmissionMay 23, 2023
18 Jun 2023 Decision Letter - Talib Al-Ameri, Editor PONE-D-23-15843A machine learning approach to model the impact of line edge roughness on gate-all-around nanowire FETs while reducing the carbon footprintPLOS ONE Dear Dr. Loureiro, Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. Please submit your revised manuscript by Aug 02 2023 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. Please include the following items when submitting your revised manuscript: A rebuttal letter that responds to each point raised by the reviewer. You should upload this letter as a separate file labeled 'Response to Reviewer'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter. If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols. We look forward to receiving your revised manuscript. Kind regards, Talib Al-Ameri, Ph.D Academic Editor PLOS ONE Journal Requirements: When submitting your revision, we need you to address these additional requirements. 1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf 2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match. When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section. 3. Thank you for stating the following financial disclosure: "Work supported by the Spanish MICINN, Xunta de Galicia, and FEDER Funds under Grants RYC-2017-23312, PID2019-104834GB-I00, PLEC2021-007662, ED431F 2020/008 and ED431C 2022/16." Please state what role the funders took in the study. If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript." If this statement is not correct you must amend it as needed. Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf. 4. Thank you for stating the following in the Acknowledgments Section of your manuscript: "Work supported by the Spanish MICINN, Xunta de Galicia, and FEDER Funds under 332 Grants RYC-2017-23312, PID2019-104834GB-I00, PLEC2021-007662, ED431F 2020/008 333 and ED431C 2022/16." We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form. Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows: "Work supported by the Spanish MICINN, Xunta de Galicia, and FEDER Funds under Grants RYC-2017-23312, PID2019-104834GB-I00, PLEC2021-007662, ED431F 2020/008 and ED431C 2022/16." Please include your amended statements within your cover letter; we will change the online submission form on your behalf. 5. Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice. [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #1: Yes ******** 2. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: Yes ****** 3. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #1: Yes ****** 4. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #1: Yes ****** 5. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #1: Review PlusOne PONE-D-23-15843 A machine learning approach to model the impact of line edge roughness on gate-all- around nanowire FETs while reducing the carbon footprint The paper deals with a highly efficient optmisation of the calculation of intrinsic variability in the nanoscale multigate transistors by using a machine learning algorithm. The usage of the machine learning approach will undoubtedly make the calculation of variability much more efficient when compared to a power approach when device simulations are run repeatedly with different sets of parameters affected by the variability sources. The paper will be of a great interest to readers and I would support it publication when following comments are addressed: 1) p. 1. & 13, line 10 & 308: The right expression has to be used instead of expression ‘...advanced simulation techniques…’ for the abbreviation of TCAD. 2) p. 2, line 21: In the sentence ‘...is relevant to apply alternative techniques, such as machine learning (ML), …’ , a reference to the machine learning technique is missing. Additionally, why the ML is viewed as an alternative technique to traditional TCAD techniques? The ML is a type of computational statistic using learning from data sets while the typical TCAD techniques will be physically based simulation techniques. 3) p. 3: A schematic of doping profile illustrating significance of doping parameters as N_s/d and X_m, and delta should be added either to Fig. 1 or as a new figure. 4) p. 3: The assumed workfunction should be mentioned for the simulations of ideal 10 nm and 22 nm gate NW FETs. 5) p. 5, lie 102 – 117: The introduction of the multi-layer perceptron (MLP) networks has no reference to the method. Is this method invented by authors? If so, theoretical description of the method has to be given. If not, a reference must be given. 6) p. 6, line 134: A number is missing in ‘...For each LER-affected device, as mentioned in Section ,’. 7) p. 6, line 153: The abbreviation LBFGS must be spelled out. 8) p. 7, Table 3: The table caption should explain in details (ideally each of them) what are the parameters listed in the table for a reader to have a better understanding. 9) p. 7, line 174: Typo in FigS. 10) p. 8, Fig. : The text in the figure caption ‘(LER-affected 22 nm devices).’ does not refer well to the studied NW FET, what are the 22 nm devices? 11) p. 8: Does the times required to train the MLP model include the time to produce the data for the training? This should be clarified. 12) p. 11, line 268: A number is missing in (see Section ). 13) p. 11: How exactly is the Transfer Learning performed? More details how the TL when predicting the variability for the 10 nm gate NW FET is done when using the machine learning outcome for the 22 nm gate NW FET. For example, are hidden layers modified in some way, etc.? 14) A few comments could be added into Conclusions about applicability of the developed MLP network for the other sources of variability due to metal grains in the gate, line edge of the gate, and random dopands. ****** 6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #1: Yes: Karol Kalna ******** [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.] While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step. https://doi.org/10.1371/journal.pone.0288964.r001
Revision 1
4 Jul 2023 Author Response We would like to thank the reviewer for their constructive comments that will undoubtedly improve the quality of the manuscript. Next, we address each comment hoping that our answers will be to the reviewer's satisfaction. Reply to Reviewer’s Comments to the Author The paper deals with a highly efficient optimisation of the calculation of intrinsic variability in the nanoscale multigate transistors by using a machine learning algorithm. The usage of the machine learning approach will undoubtedly make the calculation of variability much more efficient when compared to a power approach when device simulations are run repeatedly with different sets of parameters affected by the variability sources. The paper will be of a great interest to readers and I would support it publication when following comments are addressed: 1) p. 1. & 13, line 10 & 308: The right expression has to be used instead of expression ‘...advanced simulation techniques…’ for the abbreviation of TCAD. Reply: “Advanced simulations techniques” has been changed by the right abbreviation of TCAD, i.e. “Technology Computer-Aided Design (TCAD)”. 2) p. 2, line 21: In the sentence ‘...is relevant to apply alternative techniques, such as machine learning (ML), …’ , a reference to the machine learning technique is missing. Additionally, why the ML is viewed as an alternative technique to traditional TCAD techniques? The ML is a type of computational statistic using learning from data sets while the typical TCAD techniques will be physically based simulation techniques. Reply: ML learning techniques have been applied, for some years now, for the regression and prediction of the behavior of different quantities in physical systems. The reviewer is right, this type of techniques do not replace traditional simulation, since the latter provides more complete information on the systems being analyzed. However, due to the increasing complexity of the systems, the availability of regressors that collect complex correlations between physical variables and assist in the prediction of trends in their behavior allows more refined or oriented simulations for the analysis of optimal configurations or of a certain interest. In our case, we are evaluating the feasibility of using machine learning techniques as regression tools for figures of merit of nanometer transistors. In these devices, the correlations between the different variables no longer respond to classical parameterizations and are complex to visualize and analyze. Therefore, we believe that the use of neural networks, among other regression techniques, can provide relevant information that saves significant computational time, minimizing the computational needs for the studies described in this article. In the work presented in this paper, we demonstrate that different machine learning techniques allow obtaining regression models suitable for this type of devices and, in particular, the use of perceptrons, results in the most efficient method in terms of computation time versus the quality of the predictions obtained. Therefore, we have replaced in the manuscript the expression “...is relevant to apply alternative techniques, such as machine learning (ML)” by “is relevant to apply complementary techniques, such as machine learning (ML)”. In addition, the following references related to ML have been added to the manuscript: 1. Machine-Learning-Based Compact Modeling for Sub-3-nm-Node Emerging Transistors. https://doi.org/10.3390/electronics11172761 2. Simulator acceleration and inverse design of fin field-effect transistors using machine learning. https://doi.org/10.1038/s41598-022-05111-3 3) p. 3: A schematic of doping profile illustrating significance of doping parameters as N_s/d and X_m, and delta should be added either to Fig. 1 or as a new figure. Reply: The reviewer is right, that kind of figure will help to understand the significance of the relevant doping parameters. Therefore, we have included a new Figure (Fig. 2 in the manuscript) that presents the schematic doping profile for both 22 nm and 10 nm GAA NW FETs, and we have referred to it as follows: “These devices have an uniform p-type doping in the semiconductor channel and a n-type Gaussian doping in the source/drain (s/d) regions, that is fixed to Ns/d from the s/d contacts till a point (Xm) nearby the gate region, where the doping exponentially decays (with a slope δ), as shown in Fig. 2.” 4) p. 3: The assumed workfunction should be mentioned for the simulations of ideal 10 nm and 22 nm gate NW FETs. Reply: The work function values used for both NW FETs have been included in Table 1. 5) p. 5, line 102 – 117: The introduction of the multi-layer perceptron (MLP) networks has no reference to the method. Is this method invented by authors? If so, theoretical description of the method has to be given. If not, a reference must be given. Reply: The revised manuscript now includes a new reference [23] (Haykin S. Neural Networks and Learning Machines. Prentice Hall; 2009). 6) p. 6, line 134: A number is missing in ‘...For each LER-affected device, as mentioned in Section ,’. Reply: Thank you for noticing this mistake. The sentence has been modified to “For each LER-affected device, as mentioned in the Methodology section, we run a …”. 7) p. 6, line 153: The abbreviation LBFGS must be spelled out. Reply: The abbreviation has been spelled out as follows: “LBFGS (Limited-memory Broyden-Fletcher-Goldfarb-Shanno algorithm)”. 8) p. 7, Table 3: The table caption should explain in details (ideally each of them) what are the parameters listed in the table for a reader to have a better understanding. Reply: Following the suggestion of the reviewer, now the caption of the table clarifies some of the parameters. The caption is now as follows: “Table 3. Main characteristics of the MLP network considered in this work. Note that the solver refers to the algorithm or method used to solve the optimization problem involved in training the regressor. L2 regularization adds a penalty term to the loss function during training to prevent overfitting.” 9) p. 7, line 174: Typo in FigS. Reply: The capital “S” has been replaced by “s”. 10) p. 8, Fig. : The text in the figure caption ‘(LER-affected 22 nm devices).’ does not refer well to the studied NW FET, what are the 22 nm devices? Reply: Captions of Figs. 7, 8 and 9 have been modified to indicate the studied devices as follows: “Fig 7. Predicted and actual values for the considered figures of merit using our test dataset (LER-affected 22 nm gate length GAA NW FETs).” “Fig 8. Predicted and actual values for the considered figures of merit using our test dataset (LER-affected 10 nm gate length GAA NW FETs).” “Fig 9. Impact of the training data size on the performance of the MLP models (LER-affected 22 nm gate length GAA NW FETs).” 11) p. 8: Does the times required to train the MLP model include the time to produce the data for the training? This should be clarified. Reply: The times required to train the MLP model do not include the time to generate the training data. That information is now included in the revised text (page 10) as follows: “Note that times to generate the training data are not included in these experiments. ” 12) p. 11, line 268: A number is missing in (see Section ). Reply: The text has been modified as follows: “In other words, computing the figures of merit for the 10 nm devices when using transfer learning is about 57,800x and 1,000x faster than MC and DD simulations, respectively (see Methodology section). ” 13) p. 11: How exactly is the Transfer Learning performed? More details how the TL when predicting the variability for the 10 nm gate NW FET is done when using the machine learning outcome for the 22 nm gate NW FET. For example, are hidden layers modified in some way, etc.? Reply: When training for the 10 nm devices, we will retain the values of the model's trainable parameters from the previous model (22 nm devices) and use those initially instead of starting a training process from scratch. Therefore, the hidden layers can be modified, but the convergence is clearly faster. This information was included in the revised text (page 12), as follows: “Next, we will apply a transfer learning approach to predict the figures of merit of the 10 nm devices using as starting point the trained models used for the 22 nm devices. It means that we will retain the values of the model’s trainable parameters from the previous model (22 nm devices) and use those initially instead of starting a training process from scratch.” 14) A few comments could be added into Conclusions about applicability of the developed MLP network for the other sources of variability due to metal grains in the gate, line edge of the gate, and random dopands. Reply: The following sentence has been added to Conclusions to highlight the applicability of the developed MLP network to othe sources of variability: “Finally, it is worth mentioning that the MLP architecture could also be applied (with an adequate calibration of the network hyperparameters and weights) to other relevant sources of variability affecting semiconductor devices, such as metal grain granularity, gate-edge roughness or random discrete dopants.” Attachments Attachment Submitted filename: Response_to_Reviewer.pdf https://doi.org/10.1371/journal.pone.0288964.r002
10 Jul 2023 Decision Letter - Talib Al-Ameri, Editor A machine learning approach to model the impact of line edge roughness on gate-all-around nanowire FETs while reducing the carbon footprint PONE-D-23-15843R1 Dear Dr. Loureiro, We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements. Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication. An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org. If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. Kind regards, Talib Al-Ameri, Ph.D Academic Editor PLOS ONE Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation. Reviewer #1: All comments have been addressed ******** 2. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #1: Yes ****** 3. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: Yes ****** 4. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #1: Yes ****** 5. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #1: Yes ****** 6. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #1: I am very very very very happy with the revised version of the manuscript and recommend it publication. ****** 7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #1: Yes: Karol Kalna ******** https://doi.org/10.1371/journal.pone.0288964.r003
Formally Accepted
13 Jul 2023 Acceptance Letter - Talib Al-Ameri, Editor PONE-D-23-15843R1 A machine learning approach to model the impact of line edge roughness on gate-all-around nanowire FETs while reducing the carbon footprint Dear Dr. García-Loureiro: I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department. If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org. If we can help with anything else, please email us at plosone@plos.org. Thank you for submitting your work to PLOS ONE and supporting open access. Kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Talib Al-Ameri Academic Editor PLOS ONE https://doi.org/10.1371/journal.pone.0288964.r004

Open letter on the publication of peer review reports

PLOS recognizes the benefits of transparency in the peer review process. Therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. Reviewers remain anonymous, unless they choose to reveal their names.

We encourage other journals to join us in this initiative. We hope that our action inspires the community, including researchers, research funders, and research institutions, to recognize the benefits of published peer review reports for all parts of the research system.

Learn more at ASAPbio .