Kinetic evaluation and process analysis of COD and nitrogen removal in UAASB bioreactor

Highlights

• Kinetic modeling of TKN and COD in a UAASB reactor at steady state.

• Four kinetic models were considered to evaluate the substrate removal.

• Substrate removal followed Modified Stover–Kincannon kinetic with high R² coefficient.

• The experimental activity under various F/M showed the better efficiency of the process.

Abstract

The up-flow anoxic/aerobic sludge bioreactor (UAASB) was a unique system to study the kinetic of COD and ammonium-nitrogen removal. Four kinetic models including first-order, Grau second-order, Stover–Kincannon and Monod models were considered to evaluate the substrate removal. Stover–Kincannon model was found to be the best to describe the treatment process and fitted well with data sets. The removal of COD and ammonium-nitrogen with manipulating hydraulic retention time (HRT) were evaluated and results showed the optimum HRT of 0.64 days. Alteration trend of turbidity and sludge volume index (SVI) in various amounts of F/M showed the decrease of both factors in an effluent wastewater with F/M reduction. Moreover, the study of COD removal in batch scale system indicated the removal of 99.9, 94.83, 98.5 and 99.19% at influent COD of 1500, 2000, 3000 and 3500 mg/L, respectively.

Introduction

Nitrogen, phosphorus and their derivatives are essential for normal function of aquatic ecosystems and they are considered as a critical food source for aquatic plants in the food network. Disrupting the nutrient bioavailability and discharging in excess to water bodies have brought out a wide range of ecological effects such as eutrophication [1]. It can be noted that as a result of human activities, eutrophication has been rapidly expanding. The immense growth of phytoplankton and macroalgae, oxygen depletion, loss of benthic areas and reduction of water quality are the consequences of this process [2], [3]. In order to avoid eutrophication, it is necessary to provide suitable conditions in wastewater treatment plants to eliminate nutrients from wastewater. Most researchers have focused on biological nutrient removal (BNR) as a sustainable method and cost-effective to build or upgrade wastewater treatment plants [4]. In addition, low sludge waste and solids as well as low energy consumption can be achieved by this process [5]. Activated sludge process has attracted much attention among BNR methods. It is a unique biotechnological phenomenon and combination of diverse microbial species that are responsible to remove organic matters and nutrients from wastewater. This technology has high efficiency and low operational costs, which has been used for municipal, industrial and domestic wastewaters treatment [6], [7]. The growing trend of this process is not only because of its ability to provide precise criteria but also due to its flexibility [4].

The kinetic models have currently been developed to aid environmental engineers for designing and optimizing the wastewater treatment process. The majority of kinetic models are basically nonlinear and the nonlinear regression method is used for approximation of kinetic constant values. It should be pointed out that the nonlinear model transformed into linear one for determination of empirical constants at a limited sample size [8]. Hence, kinetic models regulate and organize the experimental data in a certain framework to better control and monitor the system performance [9], [10]. Various kinetic models have been applied to study the removal of nitrogen and COD in wastewater treatment bioreactors. First-order, Grau second-order, Monod and Stover–Kincannon are popular and fundamental models to determine substrate removal rate [11], [12], [13]. The Stover–Kincannon model is beneficial for modeling of continuous and semi-continuous mesophilic bioreactors [14]. The significant point is that this model determines the output concentrations and bioreactor operation using organic loading rate. Tomar and Gupta [15] indicated the suitability of Grau second-order and Stover–Kincannon models for prediction of nitrogen removal in anammox hybrid reactor. On the other hand, the Monod model has been introduced as the most widely used model in aerobic bioreactors and the basis of activated sludge models because of its simplicity and high accuracy [16]. The Monod model has facilitated the interpretation of substrate concentration influences on substrate consumption rate and biomass growth [17].

Although the kinetic models of activated sludge were studied in details, but UAASB has not been used for treatment of synthetic wastewater with the aim of kinetic model evaluation. There is no complete information about kinetic parameters and interaction of multiple variables in this kind of reactor feeding with synthetic wastewater. In the present study, the COD and ammonia-nitrogen removal rates were estimated using kinetic models. In addition, the effects of HRT and F/M ratio on carbon and ammonium elimination were experimented.