The application of design of experiments and artificial neural networks in the evaluation of the impact of acidic conditions on cloud point extraction.

This study aimed to analyze the impact of acidic conditions on the recovery of ciprofloxacin and levofloxacin for cloud point extraction with the Design of Experiments and Artificial Neural Networks. The design included 27 experiments featuring three repetitions of the central point for both drugs. The tested parameters included Triton X-114 concentration, HCl concentration, NaCl concentration, and incubation temperature, which were coded at five levels. After extraction, samples were analyzed using HPLC, and statistical analysis was performed using Statistica software. Polynomial equations were developed to predict recovery values based on the experimental conditions. Results indicated that Triton X-114 concentration had the most significant impact on recovery for ciprofloxacin and levofloxacin, while HCl concentration also played an important role. The highest recovery for ciprofloxacin was observed for the following conditions: 8.5 % TX-114, 0.1 M HCl, temperature of incubation 45 °C, and no addition of NaCl. The highest recovery for levofloxacin was observed under the following conditions: 9 % TX-114, 1.5 M HCl, temperature of incubation 60 °C, and no addition of NaCl. The observed recoveries were 43 % and 53 % for levofloxacin and ciprofloxacin, respectively. They were in accordance with the predicted values calculated with DoE and with ANN approach, These findings validate the use of the Central Composite Design and Artificial Neural Networks for analyzing how acidic conditions affect the recovery of both drugs, allowing for detailed analysis and optimization of the influencing factors.