The development of an autonomous sensing platform for the monitoring of ammonia in water using a simplified Berthelot method

This study demonstrates that by combining a modified version of the Berthelot method with microfluidic technologies and LED based optical detection systems, a low cost monitoring system for detection of ammonia in fresh water and wastewater can be developed. The assay developed is a variation on the Berthelot method, eliminating several steps previously associated with the method to create a nontoxic and simple colorimetric assay. The previous Berthelot method required the addition of three reagents, mixed sequentially with the sample, which complicates the microfluidic system design. With the modified method, comparable results were attained using a single reagent addition step at a 1:1 v/v reagent to sample ratio, which significantly simplifies the fluidic handling requirement for integration into an autonomous sensing platform. The intense colour generated in the presence of ammonia is detected at a wavelength of 660 nm. The method allows for ammonia determination up to 12 mg/L NH4+ with a limit of detection of 0.015 mg/L NH4+. Validation was achieved by analysing split water samples by the modified method and by ion chromatography, resulting in an excellent correlation coefficient of 0.9954. The method was then implemented into a fully integrated sensing platform consisting of a sample inlet with filter, storage units for the Berthelot reagent and standards for self-calibration, pumping system which controls the transport and mixing of the sample, a microfluidic mixing and detection chip, and waste storage. The optical detection system consists of a LED light source with a photodiode detector, which enables sensitive detection of the coloured complex formed. The robustness and low cost of the microfluidic platform coupled with integrated wireless communications makes it an ideal platform for in-situ environmental monitoring. This is the first demonstration of a fully functional microfluidic platform employing this modified version of the Berthelot method