The combination of UV–vis spectrophotometry with a chemometric calibration tool based on partial least squares (PLS) has allowed us the development of a multivariate analytical method that simultaneously estimates the concentration and size of mixtures of silver nanoparticles (AgNPs) in environmental water samples. The method is based on changes in the surface plasmon resonance band (SPRB) of AgNPs when they form aggregated/assembled structures with L-cysteine (L-cys). Measurementts were performed by employed a fixed-time kinetics method that implies that the final spectra (response) are obtained by subtstracting the solutions spectra at fixed times. Optimization of experimental conditions affecting aggregation such as time, temperature, pH and concentration of aggregating substance was performed by experimental design and response surface methodologies (RSM). A multivariate calibration model using AgNPs of known diameter size ((20 ± 3), (41 ± 3), (59 ± 5) and (79 ± 7) nm) within a concentration range between 0.62 and 2.5 mg L−1 was constructed by using a mixture experimental design and PLS. The method was finally applied to estimate size and concentration of AgNPs in AgNPs-spiked river and tap water samples. Water samples were spiked with individual, binary and ternary mixtures of AgNPs of different sizes and by using two types of AgNPs: citrate-coated AgNPs (cit-AgNPs) and polyvinylpyrrolidone-coated AgNPs (PVP-AgNPs). A good correspondence was obtained between predicted values and the total amount of AgNPs added with recovery values ranged within 80–160% for the individual mixtures, 68–108% for the binary mixtures and 60–64% for the ternary mixtures of AgNPs. Finally, transmission electron microscopy (TEM) measurements were performed for those cases where discrepancies between the expected and the obtained values were observed. TEM micrographs evidenced the presence of agglomerates or aggregates of AgNPs in some of the mixtures or water tested.
Read full abstract