Professor Robert Melchers

© 2006 by World Scientific Publishing Co. Pte. Ltd. All rights reserved. The basic concepts in structural reliability analysis using probability theory are overviewed and the important effect of structural deterioration is noted. For representation of material loss due to corrosion and for pitting corrosion appropriately accurate models are required. These should account for the many variables involved in natural environments. Recently introduced probabilistic phenomenological models for marine immersion corrosion are then described both for general and for pitting corrosion. These show that the nature of the corrosion process changes from being controlled by oxygen diffusion and later by anaerobic bacterial action. This change has important implications for longer-term prediction of general corrosion loss and for maximum pit depth. A summary is given also of the effects of steel composition, water velocity, salinity, pollution and season of first immersion.

© 2016 Informa UK Limited, trading as Taylor & Francis Group. The corrosion of steel in open seawater is influenced mainly by water temperature. If nutrients critical for bacterial activity are available, it also may be influenced by microbiological activity. The combined effect is investigated here for the case of large variations both in the seawater temperature and in concentration of dissolved nutrients. A theoretical analysis is given for these effects acting in conjunction, with or without time-dependent differences. Field observations of corrosion losses for steel sheet piling exposed at three different locations at sea sites near Vladivostok, Russia, are used in conjunction with local temperature and nutrient concentration data to obtain trends for longer term corrosion. The main objective of the paper is to build a new long-term predictive model of corrosion which may be effectively used in conditions with large variable seasonal in water temperature and in nutrient availability.

© 2016 Informa UK Limited, trading as Taylor & Francis Group. The deterioration of marine steel infrastructure caused by corrosion may be influenced by a changing climate and/or pollution level which may lead to its serviceability and structural failure. However, almost all corrosion research until recently assumed time-invariant environmental conditions. A structural reliability analysis is applied here to simulate steel sheet piles in sea water conditions under a changing environment. Corrosion of marine steel infrastructure is modelled as a spatial time-dependent process including sea water temperature and sea level rise due to global warming and dissolved inorganic nitrogen concentration increase caused by pollution. The steel sheet piles are divided vertically into 70 elements to consider the spatial variability of different corrosion zones and sea level rise effects. Two limit states are considered: (i) stress of steel sheet piles reaches their yield stress and (ii) pitting corrosion perforation to provide an alert to repair or maintenance. The results show that ignoring the effects of a changing environment can underestimate structural capacity failure risks, and pollution will have a more significant effect on capacity of steel sheet piles than sea water temperature and sea level rise caused by global warming.

© 2016 Elsevier Ltd 16S rRNA gene profiling using a pipeline involving the Greengenes database revealed that bacterial populations in innermost (proximal to the steel surface) and outer regions of biofilms on carbon steel exposed 3 m below the surface at an offshore site in the Gulf of Guinea differed from one another and from seawater. There was a preponderance of gammaproteobacterial sequences, representing organisms known for hydrocarbon degradation. Total DNA from the innermost layer was 1500 times that recovered from the outermost. Stramenopiles (diatom) sequences were prevalent in the former. Rhodobacteriaceae, key biofilm formers, comprised 14.9% and 4.22% OTUs of inner and outer layers, respectively. Photosynthetic anaerobic sulfur oxidizer sequences were also prominent in the biofilms. Analysis of data using a different pipeline with Silva111 allowed detection of 0.3¿0.4% SRB in the biofilms. The high abundance of aerobic micro-algal sequences in inner biofilm suggests they are initial colonizers of carbon steel surfaces in a marine environment. This is the first time that the microbial population of the strongly attached inner layer of the biofilm on steel has been differentiated from the outer, readily removed layer. The accepted scraping removal method is obviously inadequate and the resulting microbial analysis does not offer complete information on the biofilm community structure.

© 2015 Taylor & Francis. Carbon steel pipelines are widely used for injection of sea and other waters into oil and gas wells so as to increase the rate of recovery, particularly from mature fields. Internal corrosion usually is mild. However, cases of very aggressive channelling corrosion along the bottom of the pipeline have been observed. Practical experience and anecdotal observations have attributed this to microbiologically influenced corrosion even though extensive use is made of preventative measures including biocides, oxygen scavengers, corrosion and scale inhibitors, and pipeline pigging. Interpretation of data and observations for five water injection pipelines, made available by industry, indicate that microbiologically influenced corrosion may play a part in causing channelling corrosion but that the most likely cause is under-deposit corrosion under pipe debris that settles during periods of pipeline shut-downs and low water velocity.

© 2016 Elsevier Ltd Ductility is an important limit state for the design of reinforced concrete beams. Its implementation varies considerably between design codes. This is investigated using reliability-based assessment with ductility defined by strain ratio. The modelling uncertainty for the ductility limit state typically is much greater than that for structural strength limit state. This is reflected in the corresponding reliability indices of limit state defined for ductility. Some of these could be considered unacceptably low.

© 2015 Engineers Australia. Cast iron bridge piers, often more than 100 years old, are still in service in rail and road bridges in many parts of Australia. Increasingly, the effect of corrosion on their present and future structural safety is of interest. Field investigations and observations to assess corrosion losses and pitting of the cast iron piers of four different operational railway bridges located in tidal marine exposure conditions are described, noting that direct visual examination usually is rendered difficult by immersion, marine growth and the presence of the graphitised layer. Measured corrosion losses and pit depths showed considerable variability between piers and between bridges. Evidence was found for the influence of microbiological corrosion, fostered by nitrogenous pollution. Implications for structural safety assessment are discussed and an example given of the estimation of likely future rate of (long-term) corrosion, necessary for assessment of remaining structural safe life.

© 2016 by the authors; licensee MDPI, Basel, Switzerland. The critical role of manganese sulphide (MnS) inclusions for the initiation of the short-term growth of pitting or localized corrosion of low carbon steels has long been recognized. Classical results show that pitting probability and pitting severity increases with increased sulphide concentration for low carbon steels as a result of magnesium sulphides acting as local cathodes for initiating pitting corrosion. However, the iron carbides (cementite) in steels can also act as local cathodes for initiation of pitting corrosion. Herein it is proposed that there is competition between pits for cathodic area and that this will determine the severity of pitting and general corrosion observed in extended exposures. Preliminary experimental data for immersion exposures of up to 56 days in natural seawater of three low carbon steels show, contrary to conventional wisdom, greater pit depths for the steels with lower S content. However, the pit depth results are consistent with lower C/S ratios. This is considered to support the concept of cathodic competition between C and S. It is proposed that this offers explanations for a number of other phenomena, including the thus far unexplained apparently higher reactivity of some MnS inclusions.

© 2016 Elsevier Ltd The longer-term serviceability and structural safety of steel infrastructure exposed to seawater conditions may be affected by global warming and by seawater nutrient pollution. These may affect abiotic and biotic (microbial) corrosion. A model for long-term corrosion is developed from data obtained from steel piling exposed for 33 years in a seawater harbour. The effects on corrosion losses on the structural reliability of steel sheet piling as used in harbours world-wide were investigated as a function of seawater temperature rise from global warming and of seawater nutrient pollution. The results show that structural reliability is more sensitive to likely nutrient pollution than to predicted increases in seawater temperature, noting also that global warming also could increase nutrient pollution from anthropological sources.

© 2016 Elsevier Ltd The influence of atmospheric conditions on the corrosion of steel and its associated microbial community was studied. Surface analysis revealed greater localized corrosion in steel exposed to near-ocean atmospheres with high chloride deposition compared to inland and subalpine sites. High-throughput sequencing analysis of corrosion products showed that dissimilar microbial communities and dominant species were deposited on steel in the different atmospheres. Close to the ocean, Brevundimonas diminuta were predominant whereas Clostridium and Pseudomonas species dominated for inland sites with agricultural or forestry activities. Bacillus and Enterococcus were dominant for sites close to a fertilizer plant and a sewage treatment plant, respectively. Actinobacteria species dominated at sub-alpine conditions. Results from this study indicate that microbial communities on corroding steel exposed to atmospheric conditions are the result of deposition of locally-generated aerosols. Acid-producing activity and exopolysaccharide (EPS) production was widespread and rapidly detected in microbial cultures from all the exposure sites. Sulphate-reducing bacteria were not detected in this study. These results suggest that acid production and EPS synthesis can be important mechanisms for microbial corrosion of steel under atmospheric conditions.

© 2016 Institute of Materials, Minerals and Mining. Cast iron piers of a disused 90 year old multispan railway bridge located close to the Pacific Ocean were extensively sampled for remaining wall thickness to determine corrosion loss and pit depth. From this, a corrosion loss model for the full 90 years was developed. In addition, the statistics for uncertainty in corrosion loss were obtained. Corrosion varied with elevation relative to mean water level and was negligible in the atmospheric zone, about 2¿3 mm in the immersion zone and 5¿6 mm in the splash and lower tidal zones. This variation is consistent with accelerated low water corrosion. It indicates that water pollution occurred sometime during the life of the bridge. Maximum pit depths were determined and analysed using extreme value statistics. The corrosion model for such long term exposure and the related statistical results are unique and important for assessment of remaining life of the many other cast iron structures still in existence in many parts of the world.

© 2015 Published by Elsevier Ltd. Abstract Samples of new and 70 year old pre-corroded OPC concrete were exposed for up to 48 months in 6 sewers throughout Australia. Corrosion losses at each site followed the bi-linear trend originally proposed by Wells and Melchers [1]. During an initial phase (lasting < 2 years) negligible loss of material occurs however once the surface pH = 6 losses commence and accumulate linearly at a rate that is likely to remain constant over time. Corrosion rates were found to be sensitive to humidity but insensitive to concrete alkalinity. A first pass model which predicts the rate of concrete sewer pipe corrosion from a knowledge of local average sewer gas temperature, humidity and H < inf > 2 < /inf > S concentrations was also developed. The equation predictions were in good agreement with rates determined from field observation and historical data.

© 2014, Taylor & Francis. For environmental and for economic reasons, there is increased emphasis on extending the life of existing infrastructure and to design new infrastructure for longer, safe and effective service lives. Increasing use is being made of monitoring of performance and estimation of long-term reliability and safety, allowing also for the likelihood of long-term deterioration. To obtain optimal decision outcomes, reliance should be placed not only on data but also on accumulated scientific and engineering knowledge. In engineering, this is embodied in mathematical models. Ideally, these are of good quality, calibrated to ¿real world¿ data and have prediction capabilities. Recently, developed models of this type are described for the corrosion of steel in marine environments and simplified to models suitable for engineering applications. An example is given of the prediction of the expected corrosion loss and of the likely future rate of corrosion for a mild steel structural element exposed to temperate seawater.

© 2015, NACE International. The quantification of mass loss, surface topography, depth of pitting, and localized corrosion for steels subject to marine corrosion requires the removal of rusts, preferably without causing additional mass loss, pitting, or other damage. Conventional procedures using inhibited hydrochloric acid or Clarke's solution are shown to remove short-term marine rusts but also to cause new corrosion including pitting and related rust products. Both increase with longer exposure to the cleaning solutions. Water washing with soft water left calcareous materials on the metal surface. Ultrasonic removal did not cause additional corrosion but was effective only for very light rusts. Although relatively slow, electrolytic cleaning (electrolysis) produced clean surfaces without new corrosion.

© 2015 Elsevier Ltd. The evolution of corrosion loss and maximum pit depth of copper and copper alloys exposed for long periods of time in natural and industrial environments is shown to be more consistent with a bi-modal functional form than with the classical power law. Data from several long-term exposure test programs supports this proposition. The bi-modal behaviour signals a change from mainly cathodic oxygen reduction to a subsequent transitory corrosion process that may be modelled as involving pitting under earlier copper corrosion products. Possible reasons for some data sets showing decreasing maximum pit depths with increasing exposure time are discussed.

© 2015 Elsevier Ltd. Abstract Based on the general form of Paris' law a new method is proposed for estimating the rate of cracking of metals under Hydrogen-Assisted Fatigue. It is based on relating the fatigue crack growth rate of hydrogen embrittled metal to the fatigue crack growth rate of the metal without hydrogen embrittlement. One-dimensional hydrogen diffusion is assumed. Simulations using the proposed method on steel types X52, X70-80 and X80 are in agreement with published experimental tests results. The new method obviates numerical modelling of crack propagation and much reduces the computational costs.

© 2015 Elsevier Ltd. Chains usually form the upper part of moorings systems used for maritime structures such as floating production storage and offloading (FPSO) vessels, increasingly employed in the offshore oil and gas industry particularly in very deep waters. Current design rules do not differentiate between corrosion and inter-link wear. Laboratory experiments are described to determine the rate of wear of model (i.e. small-scale) mooring chains for up to 200,000 wear cycles. Various axial loadings and specific angular displacement were used with testing under either dry or wet conditions and for un-corroded and corroded chain. The results show that tensile force has a significant but non-linear effect on the inter-link wear. The amount of wear is similar for un-corroded and for corroded chains and is lower in wet conditions.

© 2015 Robert E. Melchers. Aluminium alloys have excellent corrosion resistance to a wide variety of exposure conditions. Usually they corrode by pitting rather than by uniform corrosion. For infrastructure applications long-term corrosion behaviour is of interest. The relatively limited long-term pitting data that is available shows that maximum and average pit depths do not follow the power law function as conventionally assumed but tend to follow a bimodal trend with exposure time. This is consistent with the bimodal trends observed previously for corrosion mass loss of aluminium alloys. Most likely it is the result of the accumulation of corrosion products over the pit mouths, leading to the gradual development of localised anoxic conditions within pits. In turn this permits the development within the pits of anoxic autocatalytic conditions, consistent with established theory for pitting corrosion of aluminium. It also is consistent with observations of hydrogen evolution from pits. The implications of this for practical applications are discussed.

© 2015 by the authors. Due to their good corrosion resistance, copper and copper alloys such as 90:10 Cu-Ni are used extensively in high-quality marine and industrial piping systems and also in marine, urban, and industrial environments. Their corrosion loss and pitting behaviour tends to follow a bi-modal trend rather than the classic power law. Field data for 90:10 copper nickel immersed in natural seawater are used to explore the effect of water pollution and in particular the availability of critical nutrients for microbiologically induced corrosion. It is shown, qualitatively, that increased dissolved inorganic nitrogen increases corrosion predominantly in the second, long-term, mode of the model. Other, less pronounced, influences are salinity and dissolved oxygen concentration.

Highly polished coupons (25 by 25 by 1.5 mm) sourced from the same steel sheet were continuously immersion-exposed either to natural coastal seawater or to seawater from the same source subjected to filtration and UV irradiation to eliminate microbiologically influenced corrosion as much as possible. This was continued for 943 days (2.6 years). Dissolved oxygen levels were very similar in both environments. On average the UV-treated seawater was 2°C warmer, but all coupons exposed to it showed less localized corrosion than those exposed to natural seawater. The typical topographical difference was about 60% as measured by surface roughness parameter S a . Mass losses in UV-treated seawater were about 10% higher than in natural seawater, but after temperature correction were similar to natural seawater for the first year and tended to be lower subsequently. At all exposure periods the rusts in UV-treated seawater were less voluminous than the rusts in natural seawater. Eventually they also contained a higher proportion of magnetite. © 2014, NACE International.

Data from a variety of field exposure programs is used to quantify the effect of concentration of dissolved inorganic nitrogen (DIN) on long-term seawater immersion corrosion loss of structural steels. A linear correlation model that asymptotes the long-term part of the previously proposed bi-modal corrosion loss model is used. It allows for average seawater temperature. Model parameters and their variability are determined and reported. The model permits prediction of long-term corrosion loss in nutrient polluted waters of known average temperature. An example shows that anthropological pollution of seawater potentially is a major hazard for corrosion of steel infrastructure. © 2013 Elsevier Ltd. All rights reserved.

Natural deterioration of mild steel exposed to marine environment compromises the long-term integrity, serviceability and safety of new and existing infrastructure and increases the risk of structural failure. Welded structures are known to be prone to even higher risks as a result of adverse effects of pitting corrosion in weld-heated areas. A bi-modal model has been shown recently to be a better description for the long-term development of the maximum depth of pits. Also, the statistics of pit depth have been shown to be better represented, for long term exposures, by the Frechet extreme value distribution. Both new developments present challenges for structural reliability analysis. Herein a linearization is used to represent long-term development of pit depth. It is shown that data for maximum pit depths can be separated into those with Gumbel statistics and those for which a Frechet distribution is more appropriate. An example is given for the reliability analysis of a welded pipeline subjected to localized corrosion. The effect of random variable uncertainty is assessed using a sensitivity study. Results show the considerable influence on the probability of failure of pit diameter and the parameters describing the pitting corrosion model. © 2014 Elsevier Ltd.

When steel sheet piling suffers accelerated low water corrosion the webs and flange-web regions of individual piles often perforate first, for reasons not fully understood. To investigate this, samples of the cross-sections of typical U and Z profile sheet piling were exposed to natural seawater for 1, 2 and 3. years. They showed localized corrosion in the central region and also near the flange-web junctions. These locations were found to have more material defects and segregation and show composition differences. It is proposed these observations are linked to localized sheet pile perforation after long exposure to seawater. © 2013 Elsevier Ltd.

A wide range of literature data including in situ immersion, tidal, coastal atmospheric and industrial exposures is used to show that the trend for longer term corrosion of aluminium alloys is nearly always more consistent with a bi-modal model than with the classical power-law function. It is proposed the bi-modal characteristic results from the accumulation of corrosion products causing localised anoxic conditions. These permit a change from predominantly cathodic oxygen reduction to hydrogen ion reduction under anoxic autocatalytic conditions within pits. This mechanism is consistent with established theory for pitting corrosion in aluminium. © 2014 Elsevier Ltd.

© 2014, (publisher). All rights reserved. Unexpectedly high levels of corrosion of sheet and other steel piling were observed in UK harbours during the 1980s, and a phenomenon termed ¿accelerated low water corrosion¿ (ALWC) caused considerable concern about the structural safety of harbour quays and facilities. This paper shows that the severity of corrosion of steel piling both in the tidal zone and in the immersion zone is correlated with the concentration of dissolved inorganic nitrogen (DIN) in the local seawater. In the lower tidal zone this is consistent with the thinning of piles associated with ALWC. Previously it has been shown that in the horizontal direction the ALWC phenomenon is associated with material variability. Models are developed to allow the prediction of expected average corrosion loss both in the immersion zone and for the lower tidal zone (the ALWC effect) as a function of DIN, water temperature and exposure period. Data are presented to show the relatively low beneficial effect of changes in steel composition. It is concluded that for long-term protection against corrosion reliance must be placed on protective and preventive measures.

Accelerated low water corrosion (ALWC) is the term given to unusually high levels of corrosion immediately below the low water level in seawater exposures of steel piling. It has been associated with microbiologically influenced corrosion but conclusive evidence is lacking. Using published data for the corrosion of steel piles exposed for periods up to 27 years at various U.S. harbor and other locations, it is shown that the severity of ALWC is correlated with the concentration of dissolved inorganic nitrogen, a critical nutrient for microbiological (bacterial) activity in seawater. This was shown previously for short-term exposures and establishes that ALWC is indeed microbiologically influenced. It also provides a means of assessing the likelihood of long-term risk of occurrence of ALWC. © 2013, NACE International.

Structural reliability theory based on probability provides a systematic, logical and consistent means to make estimates of the safety of complex structural systems. However, its application currently leaves it open to criticism that its predictions do not necessarily compare with reality and with accepted risk criteria in other contexts. Herein, it is proposed that this is because applications have not dealt satisfactorily with the issues of human error and particularly human intervention in the management of loads and resistances. Probability-based models for loads can be modified to include the influence of human intervention such as through including the effect of the degree of enforcement of regulatory requirements. Similarly, the standard of enforcement of minimum standards of material quality should be added to probability-based models for resistances. This is relevant particularly for the assessment and prediction of the future reliability and safety of existing structures. © 2013 © 2013 Taylor & Francis.

Localised or pitting corrosion can be detrimental for steel pipes and containment structures, since wall perforation may cause system failure. Herein maximum pit depth quantification and its development with time are considered for samples taken from longitudinal welds on 33-year-old tubulars exposed in Newcastle Harbour. Relationships between pit depth and material metallurgy and corrosion properties were investigated by means of standard macro-etching, rest potential and zero resistance ammetry techniques. It is considered that the observed results are the result of the lack of homogeneity at the corrosion interface caused by differences in grain size, grain structure and the potential for pitting to occur preferentially along boundaries. The results are compared to measurements for longitudinal welds obtained previously on samples of API X56 Spec 5L pipe exposed in similar waters for up to 3?5 years, showing a reasonable degree of consistency between the two sets of data. The reasons for this are discussed. © 2013 Institute of Materials, Minerals and Mining.

Much concern exists about the apparently short durability of reinforced concrete structures in marine environments. However, there are many examples of structures that have survived for long periods of time with little evidence of reinforcement corrosion. Some of these were made with seawater as mixing water. Others had very little concrete cover. Detailed examination has revealed that reinforced concrete structures made with fine or coarse aggregate consisting of calcium carbonates such as limestone or seashells or with non-reactive dolomite have extended times to commencement of corrosion initiation and to active corrosion. The reasons for this are explored herein. In addition it is shown that some structures can have serious localized reinforcement corrosion without obvious exterior signs such as concrete cracking and delamination. This requires urgent research. © 2011 Taylor & Francis Group, LLC.

This paper presents an overview of the principal features of the recently developed model for the short and the long-term marine corrosion of steels. In marine environments such as immersion, tidal and coastal atmospheric, chlorides conventionally are considered to play a major role in causing corrosion loss. The actual corrosion process is considerably more complex and besides oxidation by dissolved or atmospheric oxygen there is an important influence from the metabolic products of sulphate-reducing bacteria. The model also deals with temperature, oxygen availability, pollution and various other influences. Although the fundamental corrosion processes are all electrochemical in nature, the model does not deal specifically with these. Instead the model is built around the governing corrosion-rate controlling processes. For engineering purposes the rate of loss of material is the critical issue. The main emphasis is on the behaviour of structural and other low alloy steels. The paper closes with some observations about the applicability of the model to stainless and weathering steels and its recent extension to reinforcement corrosion in concrete structures. © 2009 Taylor & Francis Group, LLC.

Models for the prediction of corrosion mostly give the misleading impression that corrosion of steel in seawater environments is a linear function of time (the 'corrosion rate'). Non-linear functions have been proposed also, These have been calibrated to aggregated data from a wide variety of sources. This produces predictions with wide uncertainty and it is difficult to make assessments of the effect of steel composition and environmental influences. Recent research has produced a model based on the fundamental characteristics of steel corrosion, including the effect of biological influences. Detailed investigations show that the process controlling the (instantaneous) rate of corrosion changes as corrosion progresses. This is represented as a sequence of phases for which fundamental theoretical justifications and mathematical relationships have been derived. To ensure the model has practical validity it has been calibrated to actual field observations rather than laboratory data. Research findings for the effect of water temperature, dissolved oxygen levels, nutrient pollution, depth, water velocity, water salinity and steel composition are reviewed briefly. Some observations about applications and research directions are given. © 2007 American Bureau of Shipping.

The effects of different marine environmental factors on the corrosion behavior of steel plates totally immersed in salt water are studied. A new corrosion wastage model is proposed, based on a non-linear time-dependent function. This model accounts for the effects of various environmental factors, including salinity, temperature, dissolved oxygen, pH and flow velocity. A numerical example is illustrated for ships trading in different routes in the Pacific Ocean.

A stress corrosion model of a glass fibre is used within the framework of the finite element theory to predict the life of a unidirectional composite. The approach proposed here seeks to construct plausible models of comparative behaviour such that data and experience gained from short-term durability tests can be extrapolated and used to predict long-term behaviour. It is based on numerical simulation. Its significance comes from the fact that it makes use of the finite element method at the microstructural level. The finite element can capture reasonably well all aspects of damage mechanisms taking place in the microstructure. © Institute of Materials Engineering Australasia Ltd 2005.

The design rules for FPSO mooring chain are empirical, derived mainly from experience in North Sea operations. This paper reports on aspects of a major industry-sponsored investigation to ascertain the adequacy of these rules for moorings in the Tropics and what new rules might be needed. The detailed outcomes of the project and the proposals for new design rules are still under discussion and embargo. However, an overview is given of the research program underpinning the results and specifically of the corrosion testing program and the full-scale wear-testing program. Also, in an associated but separately funded test program laboratory wear tests were conducted on scale-model chains. Some results from the model tests are described also.

The long-Term corrosion of copper and copper alloys, such as 90:10 Cu-Ni, is of interest for their use in domestic and industrial piping systems, marine, urban and industrial environments including atmospheric and also as part of the containment of nuclear wastes in deep geological repositories. The evolution of the mass loss and maximum pit depth with increased exposure time is reviewed using data previously published in the literature. That data is re-interpreted. It shows that the classical 'power law' is not an appropriate model for corrosion evolution with time. However the data mainly are more consistent with the bi-modal model previously shown to apply to steels of various types and also to aluminium and aluminium alloys. The reasons for this are proposed as the result of the corrosion process changing from initially predominantly rate-controlled by the cathodic oxygen reaction and the diffusion of oxygen through some of the layers of corrosion product to, in the longer term, rate- controlled by products and processes involved in cathodic hydrogen reduction and the dissociation of water. This can occur for copper alloys but not for pure copper. In both cases pitting is a crucial aspect for the changing nature of the corrosion process. This new interpretation has implications for prediction of long-Term corrosion from short-Term observations. It provides an explanation for the frustration noted in the copper corrosion literature for the inability to predict long-Term behaviour from short-Term and electrochemical testing. It permits examination and reconciliation of the well-known influence of water pollution in causing MIC and accelerated (mainly pitting) corrosion of important copper alloys such as 90:10 CuNi.

Copyright 2015, The Society of Naval Architects and Marine Engineers. Microbiologically Influenced Corrosion (MIC) is the term used to describe corrosion where the presence and activity of microorganisms accelerates the corrosion rate or modifies the mechanism of corrosion. MIC is strongly associated with the high rates of localized pitting corrosion observed on moorings installed in some offshore fields, including fields in West Africa and South East Asia. Diagnosis of MIC in the field can be problematic, and is largely reliant on visual inspection and microbiological reactivity tests, techniques that are limited in their accuracy and reliability. As a result, other forms of corrosion and wear can be easily misdiagnosed as MIC, with potential negative consequences for the subsequent management of the integrity of the mooring system. This paper presents an overview of the most recent research regarding factors influencing rates of conventional corrosion and MIC (as determined within the SCORCH JIP). It describes recent advances in the application of functional genetic and chemical analysis techniques to diagnose MIC, improve understand of the biochemical processes causing MIC, and identify pathways to interrupt and mitigate MIC on mooring components.

Of considerable interest in various industries such as aerospace is the longer term safety of aluminium alloy structures and the effect of deterioration. Corrosion of aluminium alloys occurs mainly as pitting for which uncertainty and variability issues usually have led to maximum pit depth being considered a random variable that is also a function of time. An extreme value distribution is fitted to the statistical data obtained from multiple observations. Usually, the selection of the most appropriate model is based on the claim that one or other distribution is a 'better fit' to the data. This classical approach takes no account of prior knowledge of the underlying physicochemical process(es) that drive pitting behaviour. A more sophisticated approach uses such prior understanding. Recently it was shown that the linear model implied by the 'pitting rate' is too simplistic. Instead, a bi-modal model better represents both mass-loss as a function of exposure period and the evolution of maximum pit depth with time. This leads directly the possibility that one distribution may not be suitable for the whole range of pit depth data. These concepts are illustrated with examples.

The bi-modal characteristic of general or 'uniform' corrosion, or corrosion as measured by mass loss, as a function of exposure time now has been demonstrated extensively for various steels including low alloy, chromium and weathering steels, cast irons and, more recently also for aluminium alloys and for copper alloys. Also, it usually is independent of the environment. Further, it also occurs for the trend in maximum pitting as a function of time. This wide applicability demands explanation. Previously it was argued that the bi-modal character is the result of a change, at the metal-environment interface, from predominantly oxygen-rich corrosion conditions to predominantly locally anoxic corrosion conditions. It was argued also that pitting plays an important role in this process. Herein these concepts are extended from accepted theory for pitting under nominally oxic conditions that, within pits, become anoxic and causes the pitting process to become autocatalytic but eventually to lead to cessation of pit depth growth. There are two key outcomes. The first is the observation that pits grow episodically, that is, they grow in depth to a given limit, grow sideways and possibly amalgamate or coalesce, leaving plateaus of corroded surface in which new pitting can occur. The cycle time for this is faster under already anoxic conditions than earlier when it takes time for anoxic conditions to develop. The second outcome is that mass loss as a function of time is a function both of the pit depth and thus the pitting process and the increase in the number of pits with time. A considerable body of observational evidence is available to support these mechanisms. There also are well-developed theories that already explain the component mechanisms.

For estimating the likely service life of brittle material pipes with exterior corrosion pits and subject to hydrogen embrittlement crack propagation from the pits is of interest. Similarity of pattern of fatigue cracking from these pits under pressure fluctuations to that under monotonically increasing pressure is referred to. This similarity permits the effect of hydrogen on the cracking rate to be formulated. The depth of cracks initiated from surface pits is estimated as a function of pipe service time, based on the proposed formulation for Hydrogen-Assisted Fatigue Cracking.

Increasingly climate change, including the possible rise of seawater surface temperatures is causing some concern regarding the integrity of commercial and industrial structures in marine exposed environments. Structural integrity (serviceability and safety) may be compromised by the increased rates of corrosion reactions caused by increased average temperatures. Immersed structural systems also may be prone to microbiologically influenced corrosion. In this paper, three different extrapolation techniques are used to model expected corrosion loss. They are based on the same set of short-term empirical mass loss data for 1 year exposures. The results are compared with mass loss data obtained from steel piling exposed for 33 years. It is shown that the extrapolations from short term data can be very significantly in error compared with actual observations and with better modelling approaches. The effects of temperature increase and of increased microbiological influences are considered as part of the investigation.

Atmospheric corrosivity typically is determined by deploying flat coupons and measuring mass loss over a specific period (usually a year) and then deriving a material loss rate (typically expressed as microns/year). Usually mild steel but also zinc, copper and aluminium coupons have been used. The derived corrosion loss is used to gauge the susceptibility of infrastructure, such as bridges and buildings, to corrosion. Previously it was shown that slightly different corrosion rates are obtained with coupons of different sizes even under identical exposure environments. From a practical perspective the question arises what size and shape of coupon most accurately reflects corrosion to be expected on realistic steel infrastructure in corresponding environments. Results from previous and new exposure trials are compared to estimate the effect of coupon size over a two year exposure period. Also, the effect of shape and size of round and prismatic specimens is investigated. The results show that larger coupon surface areas yield lower corrosion losses. The latter are considered more relevant for accurate estimation of likely corrosion losses for infrastructure.

The deterioration of cement lined cast iron water mains such as used extensively in cities in Australia and many other cities world-wide has been observed to be mainly caused by external pitting corrosion. Since the pipes are buried and thus cannot c be inspected visually, there is considerable uncertainty about their condition. Indirect inspection techniques also have a high degree of uncertainty. For these reasons there is interest in long-term asset management to develop realistic models for depth and area of pitting corrosion to aid estimation of the probability of failure of older pipes. Generally it is considered that the deepest pits are of most interest in relation to failure. However, pit area is another important factor, which is crucial for estimating the rate of leakage and for assessing the mechanical behaviour and failure probability of pipes. Previous studies usually have assumed that pit area is directly related to pit depth. The present study examines pit area data collected from several pipes that have experienced many years of service (up to 129 years) and describes the relationship between pit areas and pit depths.

Long-term corrosion loss data for a variety of metals, including ferrous metals, buried in soils were reported in Romanoff (1957) and have often been quoted in the literature. There have been few investigations of the long-term trending of the data. Herein it is shown that the corrosion behaviour of ferrous metals buried in the ground follows a bi-modal trend with time. This trend is observed for a variety of ferrous metals in many different natural environments. Similar trends have been reported also for other exposure environments. Importantly, it indicates that the corrosion process changes over time. This is illustrated for corrosion mass loss and corrosion penetration for a variety of data sets drawn from Romanoff (1957).

Internal corrosion of steel pipelines before use, or temporarily taken out of use, and 'parked' on the seafloor for some time (months) continues to be an issue for offshore operators. This is irrespective of whether the pipes are oil production or water injection pipelines. Most commonly, severe pitting corrosion is observed at the 6 o'clock position. The conventional wisdom is that this is the result of microbiologically influenced corrosion (MIC). Indeed high concentrations of various bacteria have been found in the rusts in these pipes. The usual response is dosing with biocides but such treatment has been found to be not always very effective. It also may have environmental consequences if released to the sea when the pipe is put back into service or is abandoned. This paper reviews the available open literature on this topic and compares with previous findings for the severe internal corrosion at the 6 o'clock position sometimes found for water injection pipelines. This earlier work shows that periods of stagnant conditions with deposition of rusts and other debris at the bottom of (near-) horizontal pipes is the most likely reason for the severe corrosion sometimes observed. MIC can contribute to that corrosion but is not necessarily the major influence. Much depends on water quality. It is proposed that there are considerable similarities in the corrosion behaviour of parked oil pipelines and water injection pipelines and that there are considerable avenues for further research and development.

Injecting seawater (and other water) into oil reservoirs to promote the rates of production is widely used in the offshore oil and gas industry. This is done using carbon steel water injection pipelines (WIPs) for which internal corrosion sometimes is a major problem. Field experience often shows severe corrosion at the bottom part of the pipelines. This is known as channelling corrosion, 6 o'clock corrosion or bottom of the line corrosion. Field observations also show the presence of deposits at the 6 o'clock position and evidence of bacterial activity and both under-deposit corrosion (UDC) and microbiologically influenced corrosion (MIC) usually are suspected. Some WIPs treated with nitrate show more severe channelling corrosion than those without nitrate addition. This is contradictory with the primary purpose of adding nitrate to try to suppress the metabolism of aggressive sulphate reducing bacteria (SRB) within the oil reservoir. To date, the mechanism of channelling corrosion is not fully understood. The present work reports an experimental pilot study of the corrosion of carbon steel specimens under mixed deposits and simulated deoxygenated seawater environment. The influence of the presence of deposits, the treatment of seawater by filtration and ultraviolet (UV) irradiation and the addition of nitrate were evaluated by exposing carbon steel specimens to different test environments. The steel specimens were recovered from experimental vessels after 180 days and 365 days of exposure. The surface topography changes of steel surfaces were examined by stereo microscopy. The development of pitting under different test environments and the relative contributions to the formation of channelling corrosion are discussed.

Copyright © 2015 by the International Society of Offshore and Polar Engineers (ISOPE). The safety of mooring lines used for maritime floating structures is of considerable importance since they may be damaged by storm waves and by material deterioration. This paper considers chain wear and reports on results of a study on the effect of water on wear rate of chains. For this purpose, a specially-designed laboratory test rig was used to wear the surfaces of inter-link zones of chain sets under static axial load. The results show that tensile force has a significant but nonlinear effect on wear rate and that wet conditions reduce the wear rate to a considerable extent.

Copyright © (2015) by the Offshore Technology Conference All rights reserved. For years operators have reported accelerated rates of corrosion in wire rope and chain in warmer waters. Mooring design codes specify corrosion endurance of these components predominantly based on experience from the colder waters of the North Sea. The scope of the SCORCH JIP was to investigate and characterize corrosion of steel chain and wire rope moorings for a wide range of operating Floating Production Units (FPUs) and Floating Production Storage and Offtake vessels (FPSOs) in warm waters. A database was compiled of detailed corrosion measurements of in-service and retired mooring chains and wire ropes from about 30 FPUs operating in warm waters off Asia, Africa and the Americas. A set of standardized procedures were developed for forensic examinations of retrieved chain and wire rope, including photogrammetry and laboratory measurements that allowed 3D reconstruction and statistical analysis of corroded surfaces. The database was complemented by destructive tensile tests of a number of samples in order to correlate the observed degradation with residual breaking load. The SCORCH JIP also investigated the impact of sea temperature, water velocity, depth and oxygenation, steel grade and chain and wire rope construction through over 750 sample and full-scale tests spanning 3.5 years at sites around Australia. Additional laboratory tests were conducted into Microbiologically Influenced Corrosion (MIC) and the combined effect of corrosion and wear of mooring chains. The SCORCH JIP has produced a series of findings with long-lasting implications for the industry, including detailed investigations of factors in corrosion such MIC, chain pitting, chain wear, wire rope blocking compound efficacy, and the effect of environmental and operating conditions. The SCORCH JIP has produced tabulated predictions for corrosion in varying temperatures, at positions in the mooring line and for nutrient levels that could encourage MIC. The predictions are underpinned by a large number of field experiments and results from operating FPUs, and are supplemented by detailed operational guidance and recommendations for maximizing mooring corrosion endurance.

Structural risk is not the same as corrosion risk. Therefore once assessment of infrastructure confirms corrosion is active and the mechanism(s) of corrosion are understood then this does not necessarily indicate an unacceptable structural risk. Asset managers, infrastructure owners and stakeholders normally indicate a required service life requirement. An implicit requirement is meeting structural and other safety requirements as defined by Australian or other Standards. Maintenance and/or remedial work activities are only supported for the asset to achieve desired functionality, serviceability, safety, operational and commercial outcomes for its required service life. The paper explores the differences between structural and corrosion risk and how each of these risk types should be considered in the assessment, management and maintenance of infrastructure to meet required service life, structural safety and risk requirements. Some examples are used to illustrate this theme.

Copyright 2014, Offshore Technology Conference. Forensic investigations on severely corroded (pitted) chains recovered from the mooring systems of two FPUs in West Africa and Indonesia are described. During the course of the SCORCH JIP, it became apparent that a similar phenomenon had been experienced by other JIP participants indicating that it may be a common problem deserving attention. The conclusion of the present investigation is that the large pits most likely can be attributed to Microbiologically Influenced Corrosion (MIC). Where pitting occurred, the pit depth growth rate was found in instances to be up to 10 times the uniform corrosion rate, with pits distributed approximately uniformly but randomly around the link and the bar circumference. Destructive tensile tests of the samples indicated residual strength that was at worst 25% below as-manufactured estimated As-new Breaking Load (ABL). The correlation of loss of strength with loss of metallic area due to uniform and pitting corrosion indicated that breaking load cannot be treated simplistically, and requires more sophisticated consideration of the pit geometry and its location with respect to areas of high stress within the link.

The number of publications relating to the good performance of older chloride-contaminated reinforced or prestressed concrete structures continues to rise. This contradicts the majority of corrosion literature concerning structural concrete, which has continuously promoted the negative effects of chlorides since the 1960's. Various standards and guidelines reflect this sentiment, specifying empirical chloride threshold limits that are linked to corrosion risk levels. These have been adopted by industry to inform maintenance and repair procedures. It has also extended to an underlying view that detrimental corrosion may initiate and progress, which may be an incorrect assumption. To address this interpretation, a brief review is provided on the historical development of ideas and research behind chloride-contaminated concrete and associated corrosion-risk, and the subsequent establishment of accepted chloride limits. Results from 2 recent investigations relating to chloride-contaminated concrete are also discussed. A summary discussion on the implications and applications of these findings concludes the paper. This paper identifies that whilst chlorides remain an important consideration in the corrosion process, greater caution is required when assessing structural concrete for corrosion risk based purely on published chloride limits.

Microbial induced corrosion (MIC) of reinforced concrete sewer piping and manholes is a significant issue in Australia and overseas costing water authorities hundreds of millions of dollars annually. It is anticipated that as the country's sewer infrastructure ages the problem will become more severe. Over the last 4 years an ARC and industry funded research project has been undertaken with the aim of building a mathematical model to predict the corrosion of concrete as a function of exposure time and environmental and operational conditions. After almost 4 years of field trials in Sydney, Melbourne and Perth sewers a detailed understanding of the evolution of the corrosion process has emerged and a phenomenological model has been developed. The present paper describes the study findings and their implication for pipe service life prediction.

External corrosion is the main reason of the deterioration of cast iron water main in the Australia Water Industry. Prediction of the life time and the maximum pit depth which causes failure can help to reduce the cost of water system maintenance. Due to the complexities of the mechanism of pitting as well as existence of numerous uncertainties, none of the conventional electrochemical approaches is capable of accurate prediction of pipe lifetime. This study, therefore, aims at investigating the maximum pit depth based on the probabilistic theory in order to assist with the prediction of the lifetime of pipelines. A probabilistic model based on the extreme value theory is suggested for this study. Real field data of pitting corrosion of a series of old underground pipelines have been gathered from some sites in Australia. Although Gumbel distribution has been widely employed for extrapolation of maximum pit depth of pipelines, analysis of these real data indicates that the variability of maximum pit depth in long- Term exposure fits the Frechet extreme value distribution. Moreover, results of this study show that for cases whose ages have been 30-50 years the maximum pit depth tends to follow one Frechet distribution, while two subsequent Frechet curves have been observed in the probabilistic distributions of data of cases which are older than 50-60 years.

Water injection is a practical and widely employed method to promote production for oil industry especially for mature oil reservoirs. For this seawater, aquifer water and produced water (water recovered from crude oil) are most commonly used. Typically the water injection pipelines (WIP) are carbon steel and are not cathodically protected. The potential for pipeline failure caused by internal corrosion is an on-going concern for the industry. Corrosion mitigation methods include deoxygenation, corrosion inhibitors, biocides, and pigging. The most serious corrosion problem is channelling corrosion, known also as 6 o'clock corrosion or bottom of line corrosion, although it does not occur in all pipelines, for reasons not fully understood. It shows as aggressive corrosion at the bottom part of pipelines and is much more severe than corrosion elsewhere in WIPs. Recent assessment and monitoring of several water injection pipelines from the North Sea continental shelf obtained either by intelligent pigging tools or on-site bio-coupons has indicated that microbiologically influenced corrosion (MIC) and under-deposit corrosion are likely to be involved in channelling corrosion. Despite occasional industrial reports and conference papers, the comprehensive interpretation of this type of corrosion is still limited. The present paper outlines an experimental pilot study of under-deposit corrosion of mild steel model pipeline coupons exposed to deoxygenated natural seawater in simulated field conditions. Identical test rigs and different test scenarios were established to ascertain the influence of the presence of deposits, of water quality, surface roughness and nitrate addition. SEM observations of changes in surface topography for the first set of samples after 12 days exposure are reported. The paper also discusses the potential reasons of the formation process of channelling corrosion in water injection pipelines and the effect of early stage corrosion for long-term trends.

Comprehensive new data for the effects of orientation, height and bold exposure versus shielding effects and detailed chronological observations and variability data for nominally identical exposures of mild steel coupons all exposed at one site is presented. Previous investigations have been reported only for different sites and this raises difficulties in making comparisons and deductions. The present results have been obtained under the one local climatic regime for all coupons, thus providing a much more controlled environment within which to compare various effects. The present paper deals with the five-year results and compares these with information available in the literature. After three years most of the boldly exposed double-sided coupons had completely corroded but many single-sided coupons remained viable. The difference in corrosion behaviour between individual coupons exposed at different heights and vertical continuous single strips of steel is described. Moreover, the difference in corrosion losses for a continuous strip and for a series of coupons oriented in different directions is described. Furthermore, variability of nominally identically exposed coupons is reported.

Aluminium alloys have excellent corrosion resistance including in seawater and marine conditions. Herein a wide range of literature data, including in-situ immersion, tidal, coastal atmospheric and industrial exposures is examined. It is shown that while the classical power-law has been assumed, widely, to be valid as a model for the progression of corrosion, in the longer term corrosion of aluminium alloys is nearly always more consistent with the bi-modal model previously proposed only for the corrosion behaviour of steels in seawater and marine conditions, including in the atmosphere. Although aluminium alloys tend to pit rather than to corrode 'uniformly', the bi-modal trend applies irrespectively of whether mass loss, loss of tensile strength or the deepest pits are considered. It is proposed the bi-modal characteristic results from the accumulation of corrosion products causing localised anoxic conditions. These permit a change from predominantly cathodic oxygen reduction to hydrogen ion reduction under anoxic autocatalytic conditions within pits. This mechanism is consistent with established theory for pitting corrosion in aluminium, such as proposed many years ago by Wranglen. It is shown that maximum pit depths for very deep pits tend to follow the Frechet rather than the Gumbel Extreme Value distribution. This is consistent with what has been observed previously for steels.

Corrosion profiles for coupons and for single strips have been published previously and tend to show a distinct difference between the two. Corrosion of a continuous strip usually results in a flatter profile than that obtained for a set of individual coupons. Herein these results are compared with the corrosion of chain links in chains exposed through the tidal zone. Steel chains are used extensively in the marine industry and are subject to the same corrosive effects as other marine steel infrastructure. The matter of interest is whether chain links act as individual coupons or form part of a single strip as in electrically connected links of a chain. If the corrosion profile is similar to that of coupons, individual links will act as isolated cathode/anode couples and will corrode faster in the tidal zone than if the corrosion loss is spread over the entire length of the chain. The present paper presents corrosion profiles derived from lengths of nominally 10 mm diameter steel chains exposed at three geographical locations and compares these profiles to profiles obtained from single strips and sets of individual coupons exposed at the same or similar locations.

The apparently premature corrosion of steel reinforcement in reinforced concrete structures exposed to marine environments continues to be a matter of concern in practice despite decades of research effort and much research funding expenditure and despite much practical experience. Herein results are reported of a long-term (10 year) study of the influence of alkalis on the corrosion initiation and progression of 6mm diam. reinforcing bars centrally embedded in 40x40x160mm concrete specimens. These were made with seawater as mixing water and with a variety of mix proportions including some with added alkalis. All were equally well compacted by vibration. After 10 years continuous exposure in a fog-room only a limited amount of corrosion initiation was observed. These results are in stark contrast with those reported in the famous paper by Shalon & Raphael (1959) who found serious corrosion of reinforcement after only 1-2 years exposure for all mixes. It is shown herein that the corrosion loss trends in those tests are similar to those of steel exposed directly to seawater. This indicates that in the 1959 tests the concrete provided little protection against reinforcement corrosion. The reasons for this are considered herein and the practical implications discussed.

The corrosion of old buried cast iron water mains is a problem for the Australian Water Industry. To better manage their assets the industry requires a way of predicting the remaining service life of their pipes. In order to do this a predictive model of long-term corrosion loss with time is required. This paper describes ongoing work being conducted at The University of Newcastle to develop corrosion loss models for cast iron pipes in soils, as a function of exposure time and soil environment. A preliminary corrosion model was calibrated in a previous work, using a limited set of data collected from recent field work. This data base was extended in the current work by including data (of maximum pit depth and some soil properties) from historical condition assessments within the Hunter Water network. Previous studies by the authors (and others) indicate the significance of soil moisture content on the long-term corrosion loss and rate. At most sites soil moisture was only measured at a single point in time (or not at all), so the long-term average value of soil moisture (required for model calibration) was uncertain. To better estimate the average long-term soil moisture content, a simple climate-soil moisture model was developed. This paper presents the collected field data, describes the method for estimating long-term moisture content and presents the updated model calibration based on the additional data.

Forensic investigations were performed on chain links recovered from two FSO mooring systems in West Africa and Indonesia. Although these links were in tropical seawater for only 7 years, they experienced extremely severe pitting corrosion not previously seen in any available records. The material presented has direct application to ensuring the integrity of mooring systems as it provides insight into the mechanical behaviour of highly corroded chain links, statistics of the observed corrosion rates in these locations and guidelines for examining chain links. Corrosion on the surface of the links was mapped using 3D photogrammetry. The observed long-term corrosion rate is significantly higher than the average corrosion rate of 0.4 mm/year given in API RP-2SK. Analysis of the elementary composition of the rusts indicates that the large pits observed were most likely associated with Microbiologically Influenced Corrosion (MIC). The obtained results represent a breakthrough in the field of corrosion of mooring chains deployed in tropical seawater, corroborating the latest models that have been developed taking into account MIC. This work was conducted under the scope of the SCORCH JIP which involves ExxonMobil, Chevron, BP, Total, Shell, BHP Billiton, ConocoPhillips, Petrobras, Statoil, BSEE, DNV, ABS, Bureau Veritas, INPEX, SOFEC, InterMoor, KBR, SBM, Franklin Offshore, Vicinay Cadenas, Bridon, Arcelormittal, Arrium and Vryhof Anchors.

Recent detailed observations of the performance of mooring chains for a floating production unit in tropical West African waters have shown severe localized corrosion (pitting) of the steel chain after only seven years of use. This paper describes the investigation of this phenomenon as part of the Joint Industry Paper (JIP) research program SCORCH (Seawater Corrosion of Rope and Chain) funded by the major oil companies, most Classification Societies and various offshore operators and manufacturers. It is shown that there is a high likelihood that the chain has been subjected to microbiology influenced corrosion (MIC) as a result of the elevated levels of water pollution in the operational area. However, despite the large localized loss of steel in corrosion pots, the breaking load shows only a relatively small reduction compared to the Minimum Breaking Load specified in design guidelines. Copyright © (2012) by the Australasian Corrosion Association.