Open Research Newcastle
Browse

Long-term under-deposit corrosion of carbon steel pipes in seawater environment

thesis
posted on 2025-05-09, 12:51 authored by Xiang Wang
Water injection is the most commonly used method to increase the yield from oil and gas reservoirs. Seawater is typically the most convenient water source, and aquifer water and produced water (recovered from crude oil) or some combination is also used. The water injection pipelines (WIP) usually are carbon steel and often are many kilometres long. Because the insides of the pipes usually are not coated, the adverse operational environment means that internal corrosion is an on-going problem for many water injection pipelines. A corrosion problem of much concern in practice is the severe internal corrosion at the lower part of water injection pipelines in near-horizontal positions, in some cases, severe metal loss threatening the integrity of the pipelines. This type of corrosion is known variously as channelling corrosion, 6 o’clock corrosion, and bottom of the line corrosion. The examination of field observations suggests both under-deposit corrosion (UDC) and microbiologically influenced corrosion (MIC) are likely to be associated with the phenomenon. However, the mechanism of this phenomenon is still not fully understood. This thesis reports the observations of long-term corrosion of model steel pipes in a pilot laboratory study aimed at improving the understanding of development of channelling corrosion in offshore water injection pipelines. Half-pipe steel specimens were exposed continuously to stagnant and simulated deoxygenated seawater in the presence of mixed deposits for up to 365 days. The relative contributions of MIC, UDC and nitrate addition to corrosion development were investigated using four different test environments. The steel specimens were recovered after 12, 180 and 365 days of exposure and the changing surface topography was examined by Scanning Electron Microscopy (SEM) and optical microscope. The evolution of corrosion products were analysed by SEM and Energy-dispersive X-ray spectroscopy (EDS). The pit depths were measured by digital Linear Variable Differential Transformer (LVDT). These techniques present detailed graphical, morphological and chemical results of the corrosion process of mild steel in presence of deposits exposed to deoxygenated seawater. The observations show the synergistic effect of MIC and under-deposit corrosion led to severe localized corrosion. Nitrate addition caused most severe localized corrosion. This is linked to the enhanced MIC and the added nitrate plays the role of a source of critical nutrient. The progression of maximum pitting depth with increased exposure period was evaluated and a preliminary extreme value analysis of variability in maximum pit depth is presented. It is found that extreme value distribution examination shows Gumbel function is not appropriate to describe all the pit depth data. Frechet distribution is a better model to deal with the variability of the deepest pits. Finally, it is proposed that the continuous propagation of broad pits with the initiation of newer pits may explain the ultimate formation of channelling corrosion seen in practical water injection pipelines. Suggestions are provided for industrial practice on controlling rust deposition and MIC. The extreme value analysis of the pit depth data is also important for predicting failure probability due to pitting corrosion. In sum, the results in this thesis have implications for the corrosion management of water injection pipelines in the offshore oil industry.

History

Year awarded

2017.0

Thesis category

  • Doctoral Degree

Degree

Doctor of Philosophy (PhD)

Supervisors

Melchers, Robert (University of Newcastle)

Language

  • en, English

College/Research Centre

Faculty of Engineering and Built Environment

School

School of Engineering

Rights statement

Copyright 2017 Xiang Wang

Usage metrics

    Theses

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC