Formula Optimization of Emulsifiers for Preparation of Multiple Emulsions Based on Artificial Neural Networks

• Published in: Journal of dispersion science and technology Vol. 29; no. 3; pp. 319 - 326
• Main Authors: Wei, Huixian, Zhong, Fang, Ma, Jianguo, Wang, Zhengwu
• Format: Journal Article
• Language: English
• Published: Philadelphia, PA Taylor & Francis Group 01.03.2008; Taylor & Francis
• Subjects: ANN; BP network; Chemistry; Colloidal state and disperse state; emulsifier; Emulsions; Emulsions. Microemulsions. Foams; Exact sciences and technology; formulation optimization; General and physical chemistry; multiple emulsions; ANN; formulation optimization; Emulsion; Emulsions; BP network; Formulation; Preparation; Emulsifier; Neural network; multiple emulsions; Optimization
• ISSN: 0193-2691; 1532-2351
• DOI: 10.1080/01932690701716010

Formulation optimization of emulsifiers for preparing multiple emulsions was performed in respect of stability by using artificial neural network (ANN) technique. Stability of multiple emulsions was expressed by the percentage of reserved emulsion volume of freshly prepared sample after centrifugation. Individual properties of multiple emulsions such as droplet size, δ, viscosity of the primary and the multiple emulsions were also considered. A back-propagation (BP) network was well trained with experimental data pairs and then used as an interpolating function to estimate the stability of emulsions of different formulations. It is found that using mixtures of Span 80 and Tween 80 with different mass ratio as both lipophilic and hydrophilic emulsifiers, multiple W/O/W emulsions can be prepared and the stability is sensitive to the mixed HLB numbers and concentration of the emulsifiers. By feeding ANN with 39 pairs of experimental data, the ANN is well trained and can predict the influences of several formulation variables to the immediate emulsions stability. The validation examination indicated that the immediate stability of the emulsions predicted by the ANN is in good agreement with measured values. ANN therefore could be a powerful tool for rapid screening emulsifier formulation. However, the long-term stability of the emulsions is not good, possibly due to the variation of the HLB number of the mixed monolayers by diffusion of emulsifier molecules, but can be greatly improved by using a polymer surfactant Arlacel P135 to replace the lipophilic emulsifier.