journal article·2024-06Open access

Crystal growth characterization of WSe2 thin film using machine learning

Isaiah A. Moses, Chengyin Wu, Wesley F. Reinhart

Abstract

Materials characterization remains a labor-intensive process, with a large amount of expert time required to post-process and analyze micrographs. As a result, machine learning has become an essential tool in materials science, including for materials characterization. In this study, we perform an in-depth analysis of the prediction of crystal coverage in WSe2 thin film atomic force microscopy (AFM) height maps with supervised regression and segmentation models. Regression models were trained from scratch and through transfer learning from a ResNet pretrained on ImageNet and MicroNet to predict monolayer crystal coverage. Models trained from scratch outperformed those using features extracted from pretrained models, but fine-tuning yielded the best performance, with an impressive 0.99 R2 value on a diverse set of held-out test micrographs. Notably, features extracted from MicroNet showed significantly better performance than those from ImageNet, but fine-tuning on ImageNet demonstrated the reverse. As the problem is natively a segmentation task, the segmentation models excelled in determining crystal coverage on image patches. However, when applied to full images rather than patches, the performance of segmentation models degraded considerably, while the regressors did not, suggesting that regression models may be more robust to scale and dimension changes compared to segmentation models. Our results demonstrate the efficacy of computer vision models for automating sample characterization in 2D materials while providing important practical considerations for their use in the development of chalcogenide thin films.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Machine Learning in Materials Science · 2D Materials and Applications · Chalcogenide Semiconductor Thin Films

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References (56)

Show source-provided references
1. I.A. Moses, W. Chengyin, W.F. Reinhart, 2023. Evaluating transfer learning strategies for WSe2 thin film micrograph analysis, List - lifetime sample tracking, https://m4-2dcc.vmhost.psu.edu/list/data/RVJkDr8j1RPU.
2. Moses · 2024
3. Recent development in 2D materials beyond graphene10.1016/j.pmatsci.2015.02.002 ↗
4. 2D materials: to graphene and beyond10.1039/C0NR00323A ↗
5. Transition metal dichalcogenides and beyond: synthesis, properties, and applications of single-and few-layer nanosheets10.1021/ar5002846 ↗
6. Electric field effect in atomically thin carbon films10.1126/science.1102896 ↗
7. Recent development of two-dimensional transition metal dichalcogenides and their applications10.1016/j.mattod.2016.10.002 ↗
8. Large-area synthesis of highly crystalline WSe2 monolayers and device applications10.1021/nn405719x ↗
9. Direct synthesis of van der Waals solids10.1021/nn5003858 ↗
10. Experimental and numerical study of the chemical composition of WSex thin films obtained by pulsed laser deposition in vacuum and in a buffer gas atmosphere10.1016/j.apsusc.2012.03.153 ↗
11. Highly scalable, atomically thin WSe2 grown via metal–organic chemical vapor deposition10.1021/nn5073286 ↗
12. Influence of carbon in metalorganic chemical vapor deposition of few-layer WSe 2 thin films10.1007/s11664-016-5033-0 ↗
13. High-mobility three-atom-thick semiconducting films with wafer-scale homogeneity10.1038/nature14417 ↗
14. Suppressing nucleation in metal–organic chemical vapor deposition of MoS2 monolayers by alkali metal halides10.1021/acs.nanolett.7b02311 ↗
15. Realizing large-scale, electronic-grade two-dimensional semiconductors10.1021/acsnano.7b07059 ↗
16. 2D materials advances: from large scale synthesis and controlled heterostructures to improved characterization techniques, defects and applications10.1088/2053-1583/3/4/042001 ↗
17. Atomic force microscopy10.1063/1.881238 ↗
18. Advances in atomic force microscopy10.1103/RevModPhys.75.949 ↗
19. Atomic force microscopy for two-dimensional materials: A tutorial review10.1016/j.optcom.2017.05.015 ↗
20. Growth-etch metal–organic chemical vapor deposition approach of WS2 atomic layers10.1021/acsnano.0c05394 ↗
21. Wafer-scale MOCVD growth of monolayer MoS 2 on sapphire and SiO 210.1007/s12274-019-2502-9 ↗
22. Epitaxial growth of wafer-scale molybdenum disulfide semiconductor single crystals on sapphire10.1038/s41565-021-00963-8 ↗
23. Monolayer MoS2 on sapphire: an azimuthal reflection high-energy electron diffraction perspective10.1088/2053-1583/abce08 ↗
24. Microstructure segmentation with deep learning encoders pre-trained on a large microscopy dataset10.1038/s41524-022-00878-5 ↗
25. Rapid and flexible segmentation of electron microscopy data using few-shot machine learning10.1038/s41524-021-00652-z ↗
26. Machine learning-based multidomain processing for texture-based image segmentation and analysis10.1063/1.5135328 ↗
27. Unet++: Redesigning skip connections to exploit multiscale features in image segmentation10.1109/TMI.2019.2959609 ↗
28. Advanced steel microstructural classification by deep learning methods10.1038/s41598-018-20037-5 ↗
29. Image processing with ImageJ
30. Gwyddion: an open-source software for SPM data analysis10.2478/s11534-011-0096-2 ↗
31. WSXM: A software for scanning probe microscopy and a tool for nanotechnology10.1063/1.2432410 ↗
32. Mountains, https://www.nanosurf.com/en/software/mountainsmap.
33. Development and application of MIPAR™: a novel software package for two-and three-dimensional microstructural characterization10.1186/2193-9772-3-10 ↗
34. A deep learning approach for molecular classification based on AFM images10.3390/nano11071658 ↗
35. Automated searching and identification of self-organized nanostructures10.1021/acs.nanolett.0c03213 ↗
36. Molecule graph reconstruction from atomic force microscope images with machine learning10.1557/s43577-022-00324-3 ↗
37. Automated structure discovery in atomic force microscopy10.1126/sciadv.aay6913 ↗
38. Quantitative analysis of MoS2 thin film micrographs with machine learning10.1016/j.matchar.2024.113701 ↗
39. Migration-enhanced metal–organic chemical vapor deposition of wafer-scale fully coalesced WS2 and WSe2 monolayers10.1021/acs.cgd.2c01134 ↗
40. Role of bilayer graphene microstructure on the nucleation of WSe2 overlayers10.1021/acsnano.2c12621 ↗
41. Large-area single-crystal TMDs growth modulated by sapphire substrate
42. Overview: Computer vision and machine learning for microstructural characterization and analysis10.1007/s11661-020-06008-4 ↗
43. A new method for classifying and segmenting material microstructure based on machine learning10.1016/j.matdes.2023.111775 ↗
44. Adaptive characterization of microstructure dataset using a two stage machine learning approach10.1016/j.commatsci.2020.109593 ↗
45. Unsupervised microstructure segmentation by mimicking metallurgists’ approach to pattern recognition10.1038/s41598-020-74935-8 ↗
46. Modelling the steel microstructure knowledge for in-silico recognition of phases using machine learning10.1016/j.matchemphys.2020.123286 ↗
47. Scholder · 2018
48. Pytorch: An imperative style, high-performance deep learning library
49. E. Bakshy, M. Balandat, K. Kashin, Open-sourcing Ax and BoTorch: New AI tools for adaptive experimentation, URL https://ai.facebook.com/blog/open-sourcing-ax-and-botorch-new-ai-tools-for-adaptive-experimentation.
50. Snoek · 2012
51. J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, L. Fei-Fei, ImageNet: A large-scale hierarchical image database, in: 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009, pp. 248–255.10.1109/CVPR.2009.5206848 ↗
52. S. Xie, R. Girshick, P. Dollár, Z. Tu, K. He, Aggregated residual transformations for deep neural networks, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 1492–1500.10.1109/CVPR.2017.634 ↗
53. J. Hu, L. Shen, G. Sun, Squeeze-and-excitation networks, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018, pp. 7132–7141.10.1109/CVPR.2018.00745 ↗
54. C. Szegedy, S. Ioffe, V. Vanhoucke, A. Alemi, Inception-v4, inception-resnet and the impact of residual connections on learning, in: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 31, no. 1, 2017.10.1609/aaai.v31i1.11231 ↗
55. Efficientnet: Rethinking model scaling for convolutional neural networks
56. U-net: Convolutional networks for biomedical image segmentation