DUGi

Assessment of online monitoring strategies for measuring N2O emissions from full-scale wastewater treatment systems

http://dugi.udg.edu/item/http:@@@@hdl.handle.net@@10256@@14353

Clark-Type nitrous oxide (N 2 O) sensors are routinely used to measure dissolved N 2 O concentrations in wastewater treatment plants (WWTPs), but have never before been applied to assess gas-phase N 2 O emissions in full-scale WWTPs. In this study, a full-scale N 2 O gas sensor was tested and validated for online gas measurements, and assessed with respect to its linearity, temperature dependence, signal saturation and drift prior to full-scale application. The sensor was linear at the concentrations tested (0-422.3, 0-50 and 0-10 ppmv N 2 O) and had a linear response up to 2750 ppmv N 2 O. An exponential correlation between temperature and sensor signal was described and predicted using a double exponential equation while the drift did not have a significant influence on the signal. The N 2 O gas sensor was used for online N 2 O monitoring in a full-scale sequencing batch reactor (SBR) treating domestic wastewater and results were compared with those obtained by a commercial online gas analyser. Emissions were successfully described by the sensor, being even more accurate than the values given by the commercial analyser at N 2 O concentrations above 500 ppmv. Data from this gas N 2 O sensor was also used to validate two models to predict N 2 O emissions from dissolved N 2 O measurements, one based on oxygen transfer rate and the other based on superficial velocity of the gas bubble. Using the first model, predictions for N 2 O emissions agreed by 98.7% with the measured by the gas sensor, while 87.0% similarity was obtained with the second model. This is the first study showing a reliable estimation of gas emissions based on dissolved N 2 O online data in a full-scale wastewater treatment facility

This study was funded by the Spanish Government (MINECO) (CTM 2011-27163 and CTM2015-66892-R), European Commission (FP7-PEOPLE-2011-CIG 303946) and the Portuguese Fundação para a Ciência e Tecnologia (PTDC/AAC-AMB/12058/2010, UID/Multi/04378/2013, PhD grant SFRH/BD/74515/2010). Spanish and Portuguese Governments are also acknowledged for Acciones Integradas (PRI-AIBPT-2011-1232) and Luso-Espanhola action E-61/12. M. The European Commission is also acknowledged through COST action ES1202 (Water 2020). M. Pijuan acknowledges the Ramon y Cajal Research fellowship (RYC-2009-04959) from the Spanish Government. We thank Dr. Mikkel Holmen Andersen (Unisense Environment, Denmark) for providing the sensors and helpful comments

Elsevier