Abstract: In this study, optimum condition of ultrasonic wave technology of washing feather fibers is optimized by response surface test. On the basis of single factor test, the dependent variable (mesophilic aerobic bacterial content of feather fibers) and independent variables (washing time, washing temperature, liquor ratio and the dosage of acid protease) were optimized. Each variable and its interaction effects on the mesophilic aerobic bacterial content were analyzed through Box-Benhnken center composite design and response surface methodology, and the mathematical regression equation prediction model was established. The optimum washing conditions as follows: the hydrogen peroxide solution 5mL/L, pH value 4.0～6.0, the washing time 20min, the washing temperature 60℃, liquor ratio 1:40 and the dosage of acid protease 3%（owf）.Under these conditions, the content of mesophilic aerobic bacteria of washed feather fibers is 3.73×103cfu/g. By means of multiple regression method, the results show that regression model was highly significant between washing time, washing temperature, the dosage of acid protease, liquor ratio and the content of mesophilic aerobic bacteria. The experiment values were in general agreement with the predicted values and the experiment data under various conditions have validated the theoretical values. Therefore, it is feasible that response surface methodology is adapted to the washing process of feather fibers.

Key words: feather fiber , ultrasonic , washing , response surface method , optimization , mesophilic aerobic bacterial