A probabilistic approach for approximation of optical and opto-electronic properties of an opto-semiconductor wafer under consideration of measuring inaccuracy and model uncertainty

• Autorzy: Stroka Stefan M. , Heumann Christian , Suhrke Fabian , Meindl Kathrin

Identyfikatory

DOI

10.24425/opelre.2023.145863

• Języki publikacji: EN

Abstrakty

EN

This paper presents a probabilistic machine learning approach to approximate wavelength values for unmeasured positions on an opto-semiconductor wafer after epitaxy. Insufficient information about optical and opto-electronic properties may lead to undetected specification violations and, consequently, to yield loss or may cause product quality issues. Collection of information is restricted because physical measuring points are expensive and in practice samples are only drawn from 120 specific positions. The purpose of the study is to reduce the risk of uncertainties caused by sampling and measuring inaccuracy and provide reliable approximations. Therefore, a Gaussian process regression is proposed which can determine a point estimation considering measuring inaccuracy and further quantify estimation uncertainty. For evaluation, the proposed method is compared with radial basis function interpolation using wavelength measurement data of 6-inch InGaN wafers. Approximations of these models are evaluated with the root mean square error. Gaussian process regression with radial basis function kernel reaches a root mean square error of 0.814 nm averaged over all wafers. A slight improvement to 0.798 nm could be achieved by using a more complex kernel combination. However, this also leads to a seven times higher computational time. The method further provides probabilistic intervals based on means and dispersions for approximated positions.

Słowa kluczowe

EN

Gaussian process regression; machine learning; uncertainty quantification; photoluminescence; opto-semiconductor wafer measuring

• Wydawca: Stowarzyszenie Elektryków Polskich ; Polska Akademia Nauk ; Wojskowa Akademia Techniczna im. Jarosława Dąbrowskiego
• Czasopismo: Opto-Electronics Review
• Rocznik: 2023
• Tom: Vol. 31, No. 2
• Strony: art. no. e145863
• Opis fizyczny: Bibliogr. 17 poz., rys., tab., wykr.

Twórcy

autor

Stroka Stefan M.

• Department of Statistics, Faculty of Mathematics, Informatics and Statistics, LMU Munich, 80539 Munich, Germany
• ams-OSRAM International GmbH, 93055 Regensburg, Germany

• https://orcid.org/0009-0007-0433-7814

autor

Heumann Christian

• Department of Statistics, Faculty of Mathematics, Informatics and Statistics, LMU Munich, 80539 Munich, Germany

• https://orcid.org/0000-0002-4718-595X

autor

Suhrke Fabian

• ams-OSRAM International GmbH, 93055 Regensburg, Germany

autor

Meindl Kathrin

• ams-OSRAM International GmbH, 93055 Regensburg, Germany

Bibliografia

• [1] van de Schoot, R. et al. Bayesian statistics and modelling. Nat. Rev. Methods Primers 1, 1-26 (2021). https://doi.org/10.1038/s43586-020-00001-2.
• [2] Myers, D. E. Spatial interpolation: an overview. Geoderma 62, 17-28 (1994). https://doi.org/10.1016/0016-7061(94)90025-6.
• [3] Mitas, L. & Mitasova, H. Spatial Interpolation. in Geographical information systems: principles, techniques, management and applications (eds. Longley, P. A., Goodchild, M. F., Maguire, D. J. & Rhind, D. W.) 482-492 (Wiley, 1999).
• [4] Rasmussen, C. E. Gaussian Processes in Machine Learning. in Advanced Lectures on Machine Learning (eds. Bousquet, O., von Luxburg, U. & Rätsch, G.) 63-71 (Springer, 2004). https://doi.org/10.1007/978-3-540-28650-9_4.
• [5] Barnes, B. M. & Henn, M.-A. Contrasting conventional and machine learning approaches to optical critical dimension measurements. Proc. SPIE 11325, 222-234 (2020). https://doi.org/10.1117/12.2551504.
• [6] Schneider, P.-I., Hammerschmidt, M., Zschiedrich, L. & Burger, S. Using Gaussian process regression for efficient parameter reconstruction. Proc. SPIE 10959, 200-207 (2019). https://doi.org/10.1117/12.2513268.
• [7] Henn, M.-A. et al. Optimizing hybrid metrology: rigorous implementation of Bayesian and combined regression. Proc. SPIE 14, 044001 (2015). https://doi.org/10.1117/1.JMM.14.4.044001.
• [8] Chen, X., Liu, S., Zhang, C. & Zhu, J. Improved measurement accuracy in optical scatterometry using fitting error interpolation based library search. Measurement 46, 2638-2646 (2013). https://doi.org/10.1016/j.measurement.2013.04.080.
• [9] Härle, V. et al. GaN-Based LEDs and Lasers on SiC. Phys. Status Solidi A 180, 5-13 (2000). https://doi.org/10.1002/1521-396X(200007)180:1<5::AID-PSSA5>3.0.CO;2-I.
• [10] Stern, M. L. & Schellenberger, M. Fully convolutional networks for chip-wise defect detection employing photoluminescence images. J. Intell. Manuf. 32, 113-126 (2021). https://doi.org/10.1007/s10845-020-01563-4.
• [11] Oliver, M. A. & Webster, R. Kriging: a method of interpolation for geographical information systems. Int. J. Geogr. Inf. Syst. 4, 313-332 (1990). https://doi.org/10.1080/02693799008941549.
• [12] Buhmann, M. D. Radial Basis Functions: Theory and Implemen-tations. (Cambridge University Press, 2003).
• [13] Pedregosa, F. et al. Scikit-learn: Machine learning in Python. J. Mach. Learn. Res. 12, 2825-2830 (2011). https://jmlr.org/papers/volume12/pedregosa11a/pedregosa11a.pdf.
• [14] Fasshauer, G. E. Meshfree Approximation Methods with MATLAB. (World Scientific, 2007).
• [15] Walker, W. E. et al. Defining uncertainty: a conceptual basis for uncertainty management in model-based decision support. Integr. Assess. 4, 5-17 (2003). https://doi.org/10.1076/iaij.4.1.5.16466.
• [16] Patel, J. K. Prediction intervals – a review. Commun. Stat. – Theory. Methods 18, 2393-2465 (1989). https://doi.org/10.1080/03610928908830043.
• [17] Duvenaud, D. Automatic Model Construction with Gaussian Processes. (University of Cambridge, 2014). https://doi.org/10.17863/CAM.14087.

Uwagi

Opracowanie rekordu ze środków MEiN, umowa nr SONP/SP/546092/2022 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2022-2023).