Feasibility of using near-infrared measurements to detect changes in water quality

This study investigates the feasibility of using near infrared (NIR) measurements in early warning systems for sources of drinking water. Recent studies show that NIR could be used to detect changes in water quality due to the presence of chemicals. This study aims to test if this technique is capable of detecting relevant concentrations (milligram to microgram per litre range) of different chemicals, comparable to concentrations found in drinking water sources. NIR spectroscopy is interesting because of its low-costs, rapid analysis time (i.e., potential for online applications) and the broad range of substances that it can potentially detect. Water samples were taken from the surface water intake ‘Brakel’ of the drinking water company Dunea and spiked at different concentrations with 5 chemicals representative of substances that can be potentially found in surface water. The measurements and analysis were done in collaboration with Wageningen Food Safety Research (WFSR, formerly RIKILT). Results showed that 2 of the 5 tested substances could be detected using the NIR at relevant concentrations. The results therefore suggest that, at least for certain types of compounds, there is a potential to detect these using NIR spectroscopy. Although the implementation of NIR to measure aqueous matrices remains challenging due to the substantial absorption of water, promising results have been obtained. Further research is thus highly recommended to better understand the potential of this technique, but implementation of NIR technology as an early warning system for detecting organics directly in surface water in µg/l level will not be possible in the nearest future. In particular the lack of commercially available instruments which can be used online and fulfil the needed specification (i.e., sample filtration prior to analysis) currently hinders the applicability of this technique for early warning (i.e., online monitoring). Yet, very similar online tools are being used in other industries. Depending on the urgency to develop this tool further and the available financing, implementation for the water sector could be possible between 1 and 5 years from now, provided that the remaining knowledge gaps raised in this report are successfully addressed.