The purpose of this study was to determine the optimum re-extraction conditions of ethanol from catechin in Korean green tea (Camellia sinensis L.) using response surface methodology (RSM). The experiments were carried out according to a five level and two variable central composite design (CCD). The two independent variables were solvent ratio to sample content (1, 4, 7, 10, 13 mL/g) and extraction temperature (-20, -10, 0, 10, $20^{\circ}C$) on the dependent variables including yield, epigallocatechin (EGC), epicathchin (EC), epigallocatechingallate (EGCG), epicatechingallate (ECG), total catechin and caffeine. ANOVA results showed that Coefficients of determination ($R^2$) of estimated models for dependent variables were ranged from 0.9054~0.9778, while $R^2$ of caffeine were estimated 0.8770. The optimum ranges for the maximized extraction including yield, EGC, EC, EGCG, ECG, caffeine and total catechin were 4.5~7.5 mL/g in ratio of ethanol to sample and $-8{\sim}8^{\circ}C$ in extraction temperature. The actual values of yield, EGC, EC, EGCG, ECG, caffeine and total catechin, respectively, at the optimized conditions were 35.02%, 13.31%, 3.978%, 19.11%, 4.29%, 5.30% and 40.68%.