Plasmonic SPR metamaterial sensor optimized via machine learning for sensitive organic chemical detection in...

The precise identification and assessment of organic chemical compounds from polluted water have become crucial for managing their environmental impact. Biosensors have numerous benefits, which include barely preparation of samples, quick assessment durations, substantial specificity and sensitivity, as well as low limits of detection. This Proposed research demonstrates a Solid Square-Ring Perfect Circle Refractive Index Biosensor (SSRPCRIB) to identify organic chemical substances like methanol (MeOH), acetone (Ace), ethanol (EtOH), and propanol (PrOH) from the wastewater reference to water (H2O) sample in 1900 nm to 2100 nm wavelength range. They achieved high sensitivity values of 1517.14 nm/RIU, 1490.06 nm/RIU, 1488.88 nm/RIU, and 1363.63 nm/RIU in the detection of Ace, EtOH, PrOH, and MeOH, respectively. The high value of the quality factor is 1415.86 nm/RIU for PrOH detection, and the figure of detection is 1046.37 for the H2O sample. The high value of the detection range is 1710.83 for PrOH, and the high value of the detection limit is 0.000095 for EtOH. The machine learning optimized achieved mean square error (MSE) value is 1.11219 × 10^-6, and the prediction score value of regression is 0.99989. This presented work elevates their durability, specificity, and affordability, which could be used for wastewater treatment techniques.