Abstract: Abstract: As a wastewater, surimi wash water is rich in nutrients, mainly protein. It means both waste of material and pollution to environment. To maximize protein recovery from surimi wash water by eco-friendly techonolgies, purify surimi wash water and increase the utilization efficiency of aquatic resources simultaneously, a combined protein-recovery technology of fractional isoelectric precipitation and chitosan ?occulation was adopted. The surimi wash water produced from Alaska pollock processing was isoelectrically precipitated fractionally in two points of pH values, which were determined by isoelectric focusing result tested in polyacrylamide gels before. And then the low-concentration supernatant protein was ?occulated using chitosan. Influences of some dynamic (flocculation time) and thermodynamic (internal factors such as dosage and deacetylation degree of chitosan, and external factors such as temperature and pH value of solution) factors on the ?occulating process were explored centrally. Thereafter, a 3-factor 3-level experiment of Box-Benhnken design was executed, so as to find out if there was any interactive influence between the factors and obtain optimal ?occulating conditions. It showed that the isoelectric points of proteins in surumi wash water of Alaska pollock were located around 4.30 and 5.50, which indicated that fractional isoelectric precipitation was feasible in the solution. Total protein recovery up to 77.70% was recorded from isoelectric precipitation fractionally at the pH values of 5.50 and 4.30, which was more efficient than single isoelectric precipitation as reported before. Herein the protein content was decreased from 7.00 to 1.56 mg/mL, and the chemical oxygen demand (COD) was cut down from 8 960 to 3 916 mg/L. Single-factor experiments in the pretreated surimi wash water of 1.56 mg/mL protein showed that, in chitosan ?occulating process, the increase of flocculation time had a positive effect before 90 min. However, after 90 min, protein flocculating recovery lowered down with time increasing. Chitosan dosage rising led to flocculating recovery increasing before 250 mg/L, almost the same as used in original surimi wash water, and then the influence of this factor became to be a little negative. As to deacetylation degree of chitosan, higher degree gave out more flocculating recovery. The pH value of solution was a key factor of the chitosan ?occulation, and proper pH value, such as 8.0, got 87.8% flocculating recovery, other than pH values of 4.0-4.5, flocculation could hardly be achieved. As an important thermodynamic factor, temperature of solution had a remarkable effect only when it was above 50℃. Finally flocculating time and deacetylation degree of chitosan were fixed at 90 min and 95% respectively, and Box-Benhnken experiment was designed with 3 factors i.e. chitosan dosage, pH value and temperature (the range was limited to <40℃ under consideration for industrial practice) of solution. According to results of response surface analysis, there were significant interactive influences between dosage and pH value, and dosage and temperature respectively. The condition was optimized as pH value of 8.04, chitosan dosage of 369 mg/L and temperature of 19.6℃. Under the condition, protein recovery up to 94.75% could be achieved, while content of residue protein was decreased to 0.37 mg/mL. Meanwhile 3 040 mg/L COD remained in the cleaned water, which indicated that there were other materials contributing to COD value.