500HN polyimide film sandwich metamaterial absorber with enhanced sensing capabilities and assisted machine learning absorption forecasting
Issued Date
2025-05-01
Resource Type
ISSN
00303992
Scopus ID
2-s2.0-85214295852
Journal Title
Optics and Laser Technology
Volume
183
Rights Holder(s)
SCOPUS
Bibliographic Citation
Optics and Laser Technology Vol.183 (2025)
Suggested Citation
Hossen M.S., Islam M.T., Kirawanich P., Hoque A., Alenezi A.M., Baharuddin M.H., Alsaif H., Soliman M.S. 500HN polyimide film sandwich metamaterial absorber with enhanced sensing capabilities and assisted machine learning absorption forecasting. Optics and Laser Technology Vol.183 (2025). doi:10.1016/j.optlastec.2024.112335 Retrieved from: https://repository.li.mahidol.ac.th/handle/123456789/102955
Title
500HN polyimide film sandwich metamaterial absorber with enhanced sensing capabilities and assisted machine learning absorption forecasting
Corresponding Author(s)
Other Contributor(s)
Abstract
The proposed research is about unleashing the absorption properties and parametric forecasting of 500HN polyimide film sandwich deposited metamaterial absorber (MMA) in THz regime. The proposed micro-structure unit cell is ultra-thin (7.9μm) and compact (60μm) at its lowest operational frequency, with multiple absorption peaks at 4.66, 5.06, 5.82, 6.59, and 6.75 THz. The proposed MMA exhibits multiple absorption peaks with absorption coefficients of 88.51%, 99.84%, 99.72%, 95.89%, and 84.95%. To analyze the proper characteristics of polyimide absorption values was observed in different MMA configuration (i.e. unit cell dimension, available substrate height, outer patch radiator). The modified meandered line configuration at the top with gold material (Au) gives this sandwich structure a good stability in terms of sensing which have been verified in TE, TM, and TEM mode (E-field, H-field, and surface current distribution). The sensing capabilities were evaluated using six liquid samples, achieving a maximum sensitivity of 1.4 THz/RIU and a figure of merit (FoM) of 185 RIU−1, outperforming existing designs. Machine learning assisted forecasting analysis in TC-40, TC-50, TC-60 for the different MMA configurations indicates the absorption values can be predicted with a good accuracy. The regression algorithm models was assessed using R2, adjusted R2, and MSE which reveal the models goodness of fit, forecasting accuracy, and generalization for MMA.