ZARITZKY Noemi Elisabet
congresos y reuniones científicas
About the use of ferric chloride for phosphorus removal from dairy processing wastewater in activated sludge reactors
ViƱa del Mar, Chile
Congreso; ICEF 10 International Congress of Engineering and Food; 2008
Ferric chloride has been used for phosphorus removal from food industry wastewaters; however, this implies the installation of additional settlers. For this reason, Fe(III) addition directly to aeration tanks of activated sludge plants has been proposed as a suitable alternative. The objective of the present work was to study the phosphorus removal from mixed liquors of activated sludge plants using ferric chloride.Phosphorus removal was evaluated in batch systems. Activated sludges were obtained from an aerobic lab-scale reactor operated at Hidraulic Retention Time= 25 h and Solid Retention Time = 21 days. The reactor was fed with a model wastewater of dairy industries. Ferric chloride concentrations up to 2.57 mM were tested; in all cases the initial phosphate concentration was about 1.29 mM. The mixture was stirred at 300 rpm during 1 min; then, the agitation was maintained at 30 - 50 rpm for 30 min to 96 hs. Soluble phosphate (P), Fe(III) concentrations and pH were determined in the supernatant after 45 min settlement.Results showed that for all the tested conditions, an initial Fe:P molar ratio higher than 1.75 allowed removing more than 95% of P. However, pH values decreased as the initial Fe:P molar ratio increased due to the Fe(III) hydrolysis. To keep the pH around a value compatible with a biological process, NaHCO3 was added to the model wastewater; when NaHCO3 = 1000 mg l-1, the pH was above 6 for all the tested Fe:P molar ratios without affecting the phosphate removal performance. The software package CHEAQS was used to calculate pH, phosphate (P) and Fe(III) equilibrium concentrations of the tested samples as a function of the initial Fe:P molar ratio. Calculations showed that P precipitation only occurs when the added Fe(III) is high enough to decrease pH below 4; however, in our experiments up to 95% of P removal occurred at higher pH. The differences between software predictions and experimental results could be due to kinetic features that calculations do not take into account. Therefore, the importance of the experimental work in reactor design must be emphasized.