Investigating the parameters affecting the stability of superparamagnetic iron oxide-loaded nanoemulsion using artificial neural networks

Abstract

Nanoemulsions are increasingly being investigated for their fascinating capability of loading both hydrophobic and hydrophilic molecules while their stability is still an issue, being affected by various factors. In this study, to evaluate the dominant factors affecting the stability of nanoemulsions, artificial neural networks (ANNs) were implemented. Nanoemulsions of almond oil in water containing oleic acid-coated superparamagnetic iron oxide nanoparticles were prepared using a mixture of Tween 80 and Span 80 as surfactant system and ethanol as a co-surfactant. The ratio of transparency of the samples at 30 min and 7 days after preparation was taken as an indication of the stability of samples. Four independent variables, namely, concentration of nanoparticle, surfactant, oil, and alcohol were investigated to find their relations with the dependent variable (i.e., transparency ratio). Using ANNs modeling, it was concluded that the stability is affected by all variables, with all variables showing reverse effect on the stability beyond an optimum amount.

Fig. 1

3D plot of TR predicted by the ANN model at low, mid range, and high values of concentration of alcohol and surfactant

Fig. 2

3D plot of TR predicted by the ANN model at low, mid range, and high values of concentration of oil and surfactant

Fig. 3

3D plot of TR predicted by the ANN model at low, mid range, and high values of concentration of alcohol and NP

Fig. 4

3D plot of TR predicted by the ANN model at low, mid range, and high values of concentration of NP and surfactant

Fig. 5

3D plot of TR predicted by the ANN model at low, mid range, and high values of concentration of alcohol and oil

Fig. 6

3D plot of TR predicted by the ANNs model at low, mid range, and high values of concentration of oil and NP