| Polyvinyl chloride (PVC) is commonly used for drinking water pipes. Although PVC is resistant to natural environmental conditions, organic solvents may attack the pipe wall causing swelling, softening, water lines failure and drinking water pollution. Permeation is linked to underground storage tanks and random accidental spills of organic solvents or fuel derivates. Although rare, several cases have been reported. This thesis was developed with the idea of increasing the understanding of permeation process, developing new laboratory methods for permeation testing, and helping pipe companies in assess about pipe selection in zones of high risk. Near infrared spectroscopy (NIRs) was selected as the methodology. NIR is already used in both organic solvents and polymer fields for discriminative and quantitative analysis of organic compounds. The objective of the first study was to track the permeation of PVC pipes by two major organic solvents (toluene and benzene) at different concentrations. The development of principal least squares (PLS) calibrations with NIR spectra and reference data provided by the Civil Engineering lab in Iowa State University gave accurate models with the following statistic values: R2 above 90, Relative performance determinant (RPD) higher than 3, and standard errors of prediction (SEP) values similar to the standard error of the laboratory (SEL). These models could predict the permeation state of studied pipes in mm of solvent moving front, weight gain or days under permeation. The second study lead to the correlation of the pipe permeation susceptibility in pure toluene in mm/h1/2 to the pipe spectra. Spectra differences from several pipe brands and sizes were modeled with locally weighted regression (LWR) and resulted in several models which can predict with accuracy (RPD above 5) the pipe susceptibility to permeation. In summary, NIR is a suitable tool for permeation studies and may contribute to the better understanding of permeation. |
