posted on 2017-02-17, 01:22authored byDeo, Ravin Navneet
Soil corrosion is a complex and multi-disciplinary problem affecting a range of engineering assets and critical civil infrastructure. Although significant efforts are being made to improve the life-time of metallic assets in the soil environment, failures continue to occur due to deterioration brought about by external corrosion. Pipelines are one such category of buried assets. As part of urban management strategies, emphasis is placed on maintaining and extending the life of these assets. Consequently, knowledge on the corrosivity of the environment in which these pipelines are embedded is considered essential. For this purpose a range of assessment methods are available, which can be conducted on either laboratory soil samples, or in-situ conditions. However, with increase in costs associated with premature failures as well as maintenance, there is an interest to deepen the scientific understanding of the pipeline corrosion phenomenon and improve existing life-time prediction models. In this regard, the present thesis has attempted to contribute to different aspects of the soil corrosion discipline, through a mixture of intensive laboratory as well as geophysical field study.
For the first time a galvanostatic pulse technique has been successfully applied to evaluate the electrical double layer properties at the metal/soil interface as well as to determine other corrosion related parameters. This has been conducted for synthetic soil samples in order to allow control over a multitude of variables. The choice of this technique is based on its ability to isolate bulk resistances from corrosion related process, as well as its ability to identify possible effects from soil capacitances. An important outcome of this study is that an experimental procedure is given, which allows consistent treatment of soils of various grain size distributions. Secondly, another detailed laboratory study has been conducted in order to illustrate the importance of integrating spectral induced polarization methods as part of normal assessment methods for identification of potentially corrosive soils. Synthetic soil samples from the aforementioned corrosion study were analyzed for their spectral responses in the frequency range 0.1 – 1000 Hz. Further information was extracted by fitting the acquired data using the Cole-Cole model. Overall it is shown that, apart from soil resistivities, the normalized chargeability parameter correlates with the polarization resistance, an important corrosion related parameter, when the soil physical properties manifest in the corrosion process. New ways of identifying potentially corrosive soils are also presented. This effectively extends the use of electrical characterizations for assessing soil corrosivity. Following these detailed laboratory studies, a field study is reported. Here, for the first time, the combined direct current resistivity and time-domain induced polarization (DC-TDIP) profiling methods are used to assess site conditions along a pipeline right-of-way. Additional insights on the field results are provided by laboratory analysis of soil samples acquired systematically along the pipeline right-of-way from depths near to the buried pipeline. Following this, a methodological framework is suggested to assist in developing further capacities for the use of DC-TDIP methods for in-situ soil corrosivity assessments. In particular a distribution of self potential gradients is observed along the pipeline right-of-way suggesting zones of active electrolysis and enhanced external corrosivity.