Development of a new analytical method for continuous on-line and in-situ monitoring in real world non-ideal environments

Reliable, continuous on-line water quality monitoring technology is becoming vital in the drive to ensure sustainable, safe supplies of freshwater resources in light of climate change, increased industrialisation and water scarcity (Schwarzenbach et al 2006; Diamond 2004; Allan et al. 2006). This is because currently, accurate and costeffective on-line monitoring of various water quality parameters has proven difficult to achieve, as direct sensor deployment often means most sensors are unavoidably exposed to a wide range of varied and extreme measurement conditions (Diamond 2004; Allan et al. 2006, Frey and Sullivan 2004). The majority of on-line monitoring technologies currently employed are based on direct adaptations of traditional laboratory-based analytical methods, which were not originally designed for continuous or field based monitoring applications (Frey and Sullivan 2004; Danszer and Currie 1998). As the calibration models are based on Gauss's theory of least squares, they have the inherent flaw requiring strictly defined physicochemical measurement conditions in order to obtain quantitative results (Danszer and Currie 1998). This is because most sensors are not entirely selective towards one specific analyte and tend to suffer cross responses from the sample matrix. As ideal measurement conditions are rarely present in the natural environment, this invalidates the operating conditions required for reliable performance and hence leads to measurement errors. Consequently, these methods require frequent calibration, maintenance, complicated sample pre-treatment and consume large quantities of reagents in order to try and maintain their reliable performance (Frey and Sullivan 2004). However, the cost associated with maintaining instruments based on this measurement principle has greatly reduced their wide spread application, especially for remote, large-scale environmental water quality monitoring in places like the European Union (Allan et al. 2006). Therefore, simple methods that can improve the reliability, accuracy and economic costs associated with on-line monitoring such as maintenance and reagent consumption would be of great benefit to industry, government and research organisations. The aim of this research was to develop a generic new analytical method specifically developed for continuous on-line monitoring in a diversified range of non-ideal measurement conditions. This is becoming increasingly important as we produce potable water from more impaired or novel water resources and ensure that traditional water resources such as rivers and catchments are not contaminated. Hence in this paper we present a new analytical method that can enable a traditional laboratorybased sensor to intelligently respond in-situ to its measurement environment, negating the need for conventional calibration, reagents, sample pre-treatment or strictly controlled measurement conditions.