Combining Response Surface Methodology And Machine Learning For Harmonic Mean Diameter Prediction And Optimization İn The Nanoparticle Biosynthesis;

Abstract

The synthesis of nanoparticles from biological sources by green synthesis method and production optimization studies are increasing in popularity today. However, the variability of biological source and environmental effects in such processes leads to different morphology and functionality in the final product. In this study, microalgae was used as a bioreduction agent in nanoparticle synthesis and analyses of the harmonic mean particle diameter of FeSO4 concentration and its ratio with microalgae medium were carried out in particle synthesis. In this two-stage study, the experimental design was carried out first, and the particle diameters obtained by data generation were developed by machine learning. The error rates at both stages were compared and improvements were recorded. As a result, a new low-cost, fast, simple and environmentally friendly approach was introduced to solve the data insufficiency problem and used in particle diameter estimation. The results obtained showed that the proposed combined strategy provides better nanoparticle size estimates than the statistical approach alone. The proposed method is applicable to a wide range of biotechnology and bioengineering applications with significant advanced knowledge. © 2024 IEEE.