N2O emissions from a single-stage partial nitritation/anammox granule-based reactor – a model based assessment
Abstract
Completely autotrophic nitrogen removal (CANR) can be obtained in single stage granule-based bioreactors. However, their environmental footprint is compromised due to elevated N2O emissions. In this study we develop and present results from a novel spatially explicit biochemical process model of granular CANR systems that predicts N2O dynamics and stripping. The model is adapted to predict pH gradients within the granule and considers different granule size distributions and their impact on N2O emissions. The model will serve as a guiding tool to define an operational window, where aeration strategies, granule size and bulk pH are optimized to retain high nitrogen removal whilst minimizing the N2O production
Full text
* METlab, Department of Environmental Engineering, Technical University of Denmark Building 115, 2800 Kongens Lyngby, Denmark ** Department of Chemical Engineering, Universidade de Santiago de Compostela, 15782, Santiago de Compostela, Spain 1 corresponding author: [email protected]; 2 presenting author 1. INTRODUCTION N2O emissions from a single-stage partial nitritation/anammox granule-based reactor – a model based assessment Martin Morset *, Borja Valverde-Pérez*,1, Jan Michael Blum*,2, Carlos Domingo- Félez*, Miguel Mauricio-Iglesias**, Barth F. Smets* References: [1] Volcke et al. Water Science and Technology 56.9 (2007): 117-125. [2] Mulder et al. FEMS microbiology ecology 16.3 (1995): 177-183. [3] Domingo-Félez et al. Environmental Science: Water Research & Technology 2.6 (2016): 923-930. [4] Lotti et al. Water research 60 (2014): 1-14. [5] Vangsgaard et al. Water Science and Technology 67.11 (2013): 2608-2615. [6] Charles et al. Applied and environmental microbiology 83.6 (2017): e02985-16. 2. MODEL DEVELOPMENT Process performance analyzed in 4 different scenarios – granule size and alkalinty loading: 3. SCENARIOS 1D model combining NDHA [3], Advection-Diffusion approximation and pH solver Concentration profiles Parameter S1 S2 S3 S4 N-load (mgNH4-N/L·d) 200 200 200 200 DObulk (mg/L) 0.6 0.6 0.6 0.6 pHbulk (-) 7.50 7.50 7.50 7.50 Radius (μm) 1000 500 100 1000 NH4:HCO3in 1.32 1.32 1.32 0.32 pH modeling: •Brent-Dekker method •Guaranteed convergence •pH affects microbial activity by •Impact on enzymatic activity and basic cellular functions •Speciation of substrates and inhibitors 10th International Conference on Biofilm Reactors PN/A based treatment has many benefits: •Lower aeration costs (63% [1]) •Lower sludge production (90% [2]) Certain operational conditions cause N2O emissions which could offset the carbon footprint of PN/A! PDE-system discretized in space Two-scale model: •CSTR with granular biomass •Bulk and granule connected via mass transfer resistance (boundary layer) •Large granules (S1) support Anammox growth and lead to NOB suppression. •Higher N2O emisions in small granules (S2). NOB grow and N- removal is halved compared to S1. •At high oxygen load (S3) Anammox are outcompeted and all ammonia is nitrified. •In S4, alkalinity limitations caused increased N2O production through the ND pathway due to more HNO2 accumulation compared to S1. Technical University of Denmark S1 S3 S2 Simulation L O2 (mgO2-N/ L·d) N2Oe/ΔTN (%) Δ TN/TNin (%) S1 398 2.1 87.8 S2 678 4.8 44.7 S3 972 0.6 (a) 0.0 S4 398 3.4 87.2 (a) Reported as N2Oe/ΔNH4 Granule pH gradient ΔpH<0.1 ACKNOWLEDGEMENTS 4. RESULTS •Excess buffer capacity and H+ consumption by Anammox [4] counteracts H+ production by AOB in granule at alkaline conditions [5] [6] Objectives: 1) Develop a model to predict N2O emissions from granular Anammox reactors 2) Define optimal operational conditions for maximum nitrogen removal and minimum N2O emissions The work has been funded by The Danish Council for Independent Research Technology and Production Sciences (FTP) (Project N2Oman, File No. 1335-00100B). •pH gradient is steep where there is net H+ production (AOB) Granules in laboratory reactor @Metlab_DTU