Avaliação da eletrocoagulação no tratamento de efluente líquido da indústria de laticínios

• Autor(a) principal: Valente, Gerson de Freitas Silva
• Data de Publicação: 2012
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: LOCUS Repositório Institucional da UFV
• Texto Completo: http://locus.ufv.br/handle/123456789/468
• topic: Eletrocoagulação; Efluente; Indústria de laticínios; Electrocoagulation; Effluent; Dairy industry; CNPQ::CIENCIAS AGRARIAS::CIENCIA E TECNOLOGIA DE ALIMENTOS::TECNOLOGIA DE ALIMENTOS

The present work had as objectives: to evaluate the application of electrocoagulation in the treatment of liquid effluent from the dairy industry using the following operating variables: construction of the electrode material (iron, zinc and aluminum), the initial pH; electrolysis time, the electrical current density and the distance between the electrodes. The model reduced chemical oxygen demand by means of artificial neural networks, and also checked the relative importance of input variables from the neural network in the calculation of chemical oxygen demand after treatment by electrocoagulation. Finally, it evaluated the operating cost, electricity cost, cost of electrode wear and reagent cost to adjust the pH of the effluent.With the use of iron electrodes it was found that electrolysis time of 15 minutes, initial pH near neutral and electrical current density of 50 A·m-2 resulted in a greater (approximately 58 %) reduction in chemical oxygen demand. With the use of aluminum electrodes it was found that 23 minutes of electrolysis time, initial pH of 5.0 and density of electrical current of 55 A·m-2 showed the best results. The variable distance between electrodes had no significant effect. For these conditions for electrocoagulation treatment COD removal efficiency was of 52 %. With the use of zinc electrodes, the best operating condition of electrical current density was about 86.2 A·m-2, electrolysis time of 43 minutes and an initial pH of the sample near 7.2. Under the best operating conditions it was possible to reduce the chemical oxygen demand by approximately 58 %. In modeling by artificial neural networks, for the best artificial neural network trained, validated and tested, the correlation coefficient between experimental and calculated values was 0.97 for the data set used for training, 0.96 for the data set used for validation and 0.96 for the data used to test the network. The evaluation of the relative importance of input variables in artificial neural network to calculate the chemical oxygen demand after treatment by electrocoagulation concluded that the main input variable was the concentration of total dissolved solids followed by chemical oxygen demand of the effluent. Using aluminum electrodes it was found that the cost of treating (cost of electricity, cost related to electrode wear and reagent costs for pH adjustment) the liquid effluent of a dairy by electrocoagulation was R$ 2.68 per cubic meter for 70 % removal of COD.