Isotopic analysis to identify N2O production pathways and to quantify its reduction in wastewater treatment

Nitrous oxide (N₂O) emissions from biological nitrogen removal dominate the carbon footprint of the wastewater treatment (WWT) sector. Understanding both the major N₂O production pathways and its reduction to dinitrogen (N₂) is essential for effective mitigation. In WWT, N₂O is mainly produced via three microbial pathways: (i) hydroxylamine oxidation, (ii) nitrifier denitrification, and (iii) heterotrophic denitrification; only the latter can also reduce N₂O to N₂. Analysis of the isotopologues, ¹⁴N¹⁵N¹⁶O, ¹⁵N¹⁴N¹⁶O, and ¹⁴N¹⁴N¹⁸O, relative to ¹⁴N¹⁴N¹⁶O, expressed in the δ notation, enables pathway identification by comparing measured signatures to reported endmember values. However, those endmember values are derived from limited pure culture or lab incubations and may not represent complex ecosystems such as those in activated sludge of WWT plants sufficiently, necessitating system-specific source signatures. This study combines isotopic analysis of produced N₂O with dedicated process control to disentangle microbial pathways and quantify N₂O reduction under realistic operating conditions. Online N₂O isotopic measurements were performed over a one-year period using off-axis integrated cavity output spectroscopy (LGR-ABB) at two 8 m³ pilot-scale sequencing batch reactors during aeration phases treating municipal wastewater (Eawag, Dübendorf, Switzerland). Results indicate no significant contribution of hydroxylamine oxidation to N₂O production, while both nitrifier and heterotrophic denitrification emit N₂O under specific process conditions and are characterized by system specific isotopic endmembers. The availability of NH₄⁺ as an electron donor is a prerequisite for nitrifier denitrification, while heterotrophic denitrification needs low dissolved oxygen (DO) concentration. Process-specific isotopic fingerprints were applied to disentangle active pathways and assess their response to pH, carbon availability, and DO. In parallel, we calculated the fraction of reduced N₂O by applying the Rayleigh equation coupled with published fractionation factors to our data. N₂O reduction decreased within individual aeration phases and varied between cycles. Preliminary results from explainable machine learning identified pH, temperature, NO₃^-, DO, and peak NO₂^- as key drivers of N₂O reduction.  This work provides a comprehensive isotopic framework for simultaneously resolving N₂O production pathways and N₂O reduction dynamics in WWT, advancing process understanding and informing operational strategies to mitigate greenhouse gas emissions in the wastewater sector.