Abstract
Fatty acids (FA) of pumpkin seed oil have a wide range of applications in treating prostate and skin cancers. The loading of these materials on magnetic nanoparticles (MNPs) constitutes a new approach to drug delivery systems. In this study, a safe and controllable process for adsorption of FAs from pumpkin seed oil on iron oxide nanoparticles has been presented. To extract and adsorb FAs from seed oil, the saponification reaction of oil was carried out along with the innovated hydrothermal method using the evaporation technique in the presence of MNPs. The impact of various parameters such as temperature, distilled water volume, ratio of FA to iron moles, the effect of adsorption process time on FA adsorption value and MNP size were investigated experimentally. Afterward, the optimum conditions to achieve high adsorption efficiency and low particle size were investigated using a novel hybrid SA-LOOCV-GRBF neural network modeling. This method estimated parameters of generalized radial basis function (GRBF) with simulated annealing (SA) optimization and leave one out cross validation (LOOCV) technique. Using LOOCV criteria and SA, the algorithm generated a structure distinct from other methods and achieved optimum parameters without gradient method. It provided a function that was useful for experimental optimization. The results of modeling showed that this new approach could successfully predict experimental results. Under optimum conditions, an estimated particle size of about 20nm was achieved using TEM analysis with spherical morphology. The estimated FA thickness on the surface of nanoparticles was about 5nm. Fourier transform infrared spectroscopy (FTIR) confirmed the successful formation of FA layers around nanoparticles. Moreover, zeta potential and vibrating sample magnetometer (VSM) showed that particles had super paramagnetic property with a net charge of −52mv, which was suitable for pharmaceutical applications.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have