Somayeh Firouzeh - Department of Chemistry, Faculty of Science, Bojnourd Branch. Islamic Azad University, Bojnourd,Iran
Malihe Samadi Kazemi - Department of Chemistry, Faculty of Science, Bojnourd Branch. Islamic Azad University, Bojnourd,Iran

The selection of appropriate conditions for increasing the extraction efficiency is important. The response surface method (RSM) is a statistical and mathematical technique which it can be used to examine the dependence between the responses with variables (dependent and interaction terms) in a process and also determination the optimal conditions [1]. In this study, the experimental design of General full factorial (GFF) was used. The variables were time (30, 60 and 90 min), solvent (water, methanol and n-hexane) and method (maceration and ultrasound-assisted extraction). The main effects and also interaction terms were investigated on amounts of total phenolic content (TPC), total flavonoid content (TFC) and the antioxidant activity (according to DPPH, FRAP, IC50 and BCB) of extract of Arenaria hispanica L. A multiple regression, first degree model was used to express the responses. An analysis of variance (ANOVA) with 95% confidence level was then carried out for each response (TPC, TFC, IC50, FRAP and BCB) in order to test the model significance and suitability. The significance of independent variables in the form of main effect and interaction terms were analyzed by computing the F-value at probability (p) of 0.001 and 0.05. Water was optimal solvent for TPC, FRAP and BCB, but methanol solvent was obtained for TFC. Also, the extraction time for TPC,TFC, IC50, FRAP and BCB responses was 60, 30, 60, 90 and 30 min; respectively. The maceration was selected for all responses except of TFC. The mounts of TPC,TFC, IC50, FRAP and BCB in the optimum conditions obtained 173.93±0.50(mg GA/g extract), 184.27±4.04(mg QC/g extract), 0.69±0.03 (mg/mL), 734.80±2.02(mmol Fe2+/g exctract) and 21.23±0.22; respectively. Also, the experimental values were in close agreement with values predicated by model.