Journal of Iranian Ceramic Society
Volume 20, Issue 3 (11-2024)
Jicers 2024, 20(3): 0-0 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Ghanbari H, Maleki M, Maroufi A. Examining the Effects of Factors on the Synthesis of MoS2 Nanosheets by chemical vapor deposition method using machine learning approach. Jicers 2024; 20 (3)
URL: http://jicers.ir/article-1-521-en.html
URL: http://jicers.ir/article-1-521-en.html
Iran university of science & Technology , hajar_ghanbari@iust.ac.ir
Abstract: (21 Views)
The synthesis of MoS2 nanosheets, as a strategic two-dimensional material in the fields of electronics and nanotechnology, presents a significant challenge in materials engineering. Conventional methods for synthesizing MoS2 often face limitations such as scalability, process complexities, and numerous challenges in achieving optimal conditions. In this research, novel methods utilizing machine learning approaches have been employed to investigate the effects of key factors on the synthesis of MoS2 nanosheets via chemical vapor deposition (CVD). This study incorporates statistical and deep learning techniques, along with datasets prepared from reliable sources and data simulation and generation to address the issue of limited experimental data. Supervised learning models were used for modeling on training samples. To enhance the interpretability of synthesis factors, the SHAP (SHapley Additive exPlanations) library was applied within the neural network model. The results revealed that the synthesis factors with the most significant impact on the thickness of the synthesized nanosheets, after the learning and adaptation processes with supervised learning models, are the furnace temperature, followed by the carrier gas flow rate, and to a lesser extent, the durability time during thermal operations. Approximately 1200 simulated data points were used to achieve the optimal model for predicting the thickness of MoS2 nanosheet layers. The parameters of the classification algorithms were evaluated using a combination of four metrics: accuracy, precision, F1 score, and recall. The output of the designed model, along with various hyperparameter tuning methods, was used to optimize the parameters of the machine learning models, achieving an accuracy of approximately 95% based on different classification algorithm metrics (precision/recall).
Keywords: Chemical Vapor Deposition, Machine Learning, Optimization, Energy Storage, Molybdenum Disulfide
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |