Optimization of nitrogen removal from an anaerobic digester effluent by electrocoagulation process. Issue 3 (June 2019)

• Record Type: Journal Article
• Authors: Mohammadi, Azam; Khadir, Ali; M.A.Tehrani, Ramin
• Is Part Of: Journal of environmental chemical engineering. Volume 7:Issue 3(2019)
• Publication Date: 2019-06
• Subjects: Nitrogen -- Electrocoagulation -- Digester -- Effluent -- Anaerobic -- Adsorption; Chemical engineering -- Environmental aspects -- Periodicals; Environmental engineering -- Periodicals; Chemical engineering -- Environmental aspects; Environmental engineering; Periodicals; 660.0286
• DOI: 10.1016/j.jece.2019.103195
• Languages: English
• ISSNs: 2213-2929

Abstract

Graphical abstract: Highlights: Nitrogen was removed from anaerobic digester supernatant by electrocoagulation. Iron proved more suitable electrode for nitrogen removal. N2 gas generation and precipitation of the N-species are proposed as the main removal mechanism. Among two and three-parameter isotherms, Freundlich and Sips indicated a better fitting. Abstract: The present study deals with the applicability of electrocoagulation to remove total nitrogen from actual anaerobic digester effluent for the first time. The effects of the main operational parameters including electrolysis time, current density, inter-electrode distance, agitation speed, electrode material and punched electrodes were investigated. The results showed that at 100 min of electrolysis time, 66.32% of nitrogen was removed. Current density of 0.9 A revealed a higher nitrogen elimination. At inter-electrode distance of 3 and 7 cm, maximum efficiency and energy consumption was achieved, respectively. In terms of electrode materials, the maximum and the minimum nitrogen removal efficiency of 53.4% and 42% was attributed to Fe/Al, and Al/Al, respectively. Under optimum conditions, 81.59% of total nitrogen was removed. The mechanism of electrocoagulation was modeled using removal kinetics, adsorption kinetics, two and three-parameter isotherms. Linear and non-linear method was investigated and the best model was chosen according to different error functions. Pseudo first order model describes the reaction well since R 2 (0.9997) value was too close to unity, and also error function values including Δq(%), SAE, ARED, and χ 2 were 1.7, 48.84, 3.11, and 4.33, respectively. In terms of R 2 value, it exhibited that Freundlich (0.953) and Sips (0.952) models were suitable to describe adsorption, suggesting heterogeneous adsorbent and quasi-Gaussian distribution energy.