DETERMINATION OF AMILORIDE HYDROCHLORIDE AND FUROSEMIDE IN COMPOUND FUROSEMIDE TABLETS BY CAPILLARY ELECTROPHORESIS COMBINED WITH RESPONSE SURFACE METHODOLOGY AND ARTIFICIAL NEURAL NETWORK

摘要

A capillary electrophoresis (CE) method was proposed with the application of response surface methodology (RSM) and artificial neural network (ANN) for the determination ofamiloride hydrochloride (AM) and furosemide (FUR) in Compound Furosemide Tablets. Box-Behnken design (BBD) of RSM was used for the optimization of CE experimental conditions. Statistical models were constructed by Box-Behnken design as well as artificial neural network using the three selected variables (buffer concentration, buffer pH, and voltage). Optimal separation conditions obtained were 16 mM NaH2PC>4 buffer, pH 7.21, 24 kV applied voltage, 6 s hydrodynamic injection at 30 mbar, and UVdetection at 223 nm. Eleclrophorelic analysis xoas completed in less than 5min, xoith LODs of 0.31 ng/mLforAM and 0.66 [igl mL for FUR. The method was successfully applied to determine AM and FUR in Compound. Furosemide Tablets with RSD lower than 2.2Ya, and the recovery from 98.^/a to 102.5%. Comparative experiments were also carried out using HPLC method described in Chinese Pharmacopoeia. The validation results of the tioo methods are comparable, but the analysis lime and reagent consumption for the proposed CE method were decreased significantly.