Abstract
Many metal ions are known to have severe health and environmental impacts, particularly in aqueous environments. Therefore, we have investigated the detection of divalent metal cations in water using array sensing techniques. Array sensing, also known as chemical fingerprinting or pattern recognition, can offer rapid detection of metal cations using simpler instrumentation that can be conducted on-site immediately following sample collection. Through our research group’s recent involvement in a MS-AL EPSCoR consortium, we intend to apply these results to metal cation detection and water quality monitoring along the Gulf Coast. A multivariate data set obtained by exposing an array containing two water soluble commercially available dyes (xylenol orange and methylthymol blue) to multiple metal ions was interpreted using pattern recognition algorithms such as linear discriminant analysis (LDA). The array was shown to discriminate nine divalent metal cations qualitatively with excellent resolution. We optimized our array further by taking advantage of responses from a variety of metal-dye stoichiometries to map metal ion concentrations of four environmentally relevant divalent metal ions (Hg(II), Pb(II), Cd(II), and Cu(II)) quantitatively at concentrations as low as 1 µM in neutral water. Based on each metal cation’s strong binding affinity to both dyes, near stoichiometric binding, and the system's overall linear response over a wide range of metal ion concentrations, our array was used to successfully discriminate binary and ternary metal cation mixtures, a particularly valuable accomplishment for chemical fingerprinting systems, which otherwise typically struggle with the “problem of mixtures”. To validate our array’s utility as a calibration plot, we projected a series of unknown metal cation mixtures on our original scores plot with exceptional predictive ability.
