This study focuses on determination and prediction of Phase Inversion Temperature (PIT), as a low energy method for nanoemulsion formation, using laboratory methods and artificial intelligence, respectively. A solution of lemon oil in water and tween 40, as surfactant, was prepared. For laboratory determination of PIT and scrutinizing of the effects of concentration of NaCl in aqueous phase and weight ratio of surfactant/oil (SOR) on the PIT, both visual and electrical conductivity methods were used. According to the laboratory results, the PIT decreases by increase in NaCl concentration and, it varies by variation in SOR. Because the laboratory methods are very expensive and time consuming, Adaptive Neuro-Fuzzy Inference System (ANFIS), which is one of the efficient techniques of artificial intelligence, was used for prediction of PIT. A total of 30 data sets of the studied solution, including PIT and salt concentration in aqueous phase and SOR, were used. The data are the same that obtained from the laboratory methods. These data were divided into two groups; one group included 25 data sets used for constructing the model, and the other included 5 data sets used for the correlation testing. The measured mean squared errors (MSEs) of predicted PIT from the intelligent model in the test data were 0.239 and correlation coefficient (R2) between predicted values from the model and experimental values in the test data were 0.995. The results show the model could be applied as a very fast and cheap method for PIT estimation before laboratory testing.

Username
Password