. In the water and wastewater treatment industry, coagulation is a common process in removing turbidity in the form of suspended and colloidal materials. Many coagulants are widely used in conventional water treatment processes including inorganic and polymeric coagulants. In recent years because of problems associated with usage of these coagulants including cost and health related issues, there has been considerable interest in the development of usage of natural coagulants which can be produced or extracted from microorganisms, animal or plant tissues. The advantages of these coagulants over the synthetic ones include safety to human health, biodegradability and a wider effective dose range for various colloidal suspensions. In this study, ability of Oak seed as a coagulant and coagulant aid was evaluated. To achieve this goal tannic acid (one of the main constituents of oak seed) was used. In this study two kinds of tannic acid (laboratory tannic acid and industrial tannic acid) and two kinds of turbid samples (Zayanderood river water with natural turbidity and synthetic turbid water) were used. In order to optimize the removal efficiency, seven factors including pH, coagulant dosage, rapid mixing rate, slow mixing rate, slow mixing time, sedimentation time and initial turbidity were investigated and experiments were carried out in batch reactors and on four samples: Lab. Tannic acid-Syn. Turbidity, Lab. Tannic acid-Natural Turbidity, Ind. Tannic acid-Syn. Turbidity and Ind. Tannic acid-Natural Turbidity. Taguchi method was used in order to decrease the number of experiments, expenses and time. Using an L32 array, optimum conditions for each of the samples were determined. Results showed that in optimum conditions laboratory tannic acid is capable of reducing turbidity with 60/7% and 51/9% efficiency in synthetic and natural turbid samples while industrial tannic acid yielded 49/4% and 46/9% turbidity removal efficiency for synthetic and natural turbid samples. ANOVA showed that all of the experimental factors were statistically significant in Ind. Tannic acid-Syn. Turbidity and Ind. Tannic acid-Natural Turbidity samples and in Lab. Tannic acid-Syn. Turbidity and Lab. Tannic acid-Natural Turbidity samples coagulant dosage was found to be the only statistically insignificant parameter. ANOVA results revealed that the contribution of error term in the total results for all the samples except Lab. Tannic acid-Natural Turbidity sample is less than 16% and for this sample is equal to 18%. Interaction analysis were carried out to investigate the relative importance of different interactions between involved factors in this sample and it was concluded that the most important interaction in this sample happens between rapid mixing rate and slow mixing rate.
