Search

Picric acid (PA) is a severe environmental and security risk due to its unstable, toxic, and explosive properties. It is also challenging to detect in trace amounts and in situ because of its highly acidic and anionic character. Here, we assess sensing of PA under nonlaboratory conditions using surface-enhanced Raman scattering (SERS) silver nanopillar substrates and hand-held Raman spectroscopy equipment. The advancing elasto-capillarity effects are explained by molecular dynamics simulations. We obtain a SERS PA detection limit on the order of 20 ppt, corresponding attomole amounts, which together with the simple analysis methodology demonstrates that the presented approach is highly competitive for ultrasensitive analysis in the field.

This thesis is concerned with measurements of persistent organic pollutants (POPs) in environmental samples. The POPs consist of several groups of compound with many isomers, hence analysing POPs in a complex environment is a This thesis is concerned with measurements of persistent organic pollutants (POPs) in environmental samples. The POPs consist of several groups of compound with many isomers, hence analysing POPs in a complex environment is a challenging task. Studies of the abundance or dynamics of such substances in the environment require careful consideration of several key steps. The environmental and analytical problems must be correctly defined, suitable analytes, sampling strategies and techniques must be selected, the chemical analyses must be reproducible and accurate, and the data evaluation protocols must be rigorous and appropriate. The aim of this thesis was to evaluate the use of pollutant profiles in order to assess complex patterns of environmental exposure. When combined with multivariate data analysis (MVDA) a maximum of information may be extracted from analytical data. Nevertheless, good quality data is essential for correct evaluation of the environmental phenomena under investigation. The precision and accuracy of a solid phase assisted liquid extraction method was therefore assessed by comparison to a reference method. The validated method was used in two human exposure studies. The concentrations of PCBs and dioxins were determined in human tissues from general populations in Sweden and Spain. Although it was concluded that the populations were exposed to similar degree to PCBs and dioxins, principal component analysis (PCA) showed that the compound profiles differed between the populations. PCA was also used to differentiate between occupationally exposed workers and matched controls and between exposed workers with recent and earlier exposure. In addition, the analysis indicated that the differences in PCB patterns in workers with recent and earlier exposure were related to differences in the metabolic degradation rates of individual PCBs. The PCA model was further used to select PCB congeners that were representative of each group. Finally, the emission of polycyclic aromatic hydrocarbons (PAHs) in a traffic tunnel was estimated by measuring 29 individual PAHs. The sampling periods were chosen to reflect a maximum variation in the traffic composition. An excellent prediction of the percentage of heavy-duty vehicles was obtained using partial least squares (PLS) regression. It was shown by PCA that it might be possible to predict source compositions at other locations with the aid of multivariate statistical tools. Thus, properly gathered pollutant profiles and MVDA may be combined to extract a large amount of significant information from environmental source and exposure data.