640 publications from this institution
Resistance deterioration which is purely a time function such as typified by simple corrosion problems is considered herein for structural systems subjected to one or more stochastic load processes. The resistances are modelled as Gaussian random variables and the loads as Gaussian random processes. Deterioration is considered by either of two simple models -exponential or parabolic functions of time. Both an upcrossing and an outcrossing formulation is used to estimate the up (out) crossing rate through linear domain boundaries. Numerical integration is employed to evaluate the resulting integrals, although significant simplifications are occasioned through the use of Gaussian random variables and processes. Results are given for two simple structural examples and compared to purely simulation results both for the outcrossing rate and for various sensitivity results. Good agreements is found. The results suggest that results for failure probability based on mean values of the deterioration models may be satisfactory in some cases.
The next section outlines the recent development of an appropriate model for fn(t, E) based on actual and very detailed observations of steels exposed to natural seawater. The model has been shown to be applicable to immersion, tidal and coastal atmospheric corrosion environments (Melchers 2003a, 2007b). Next is described how the model may be slightly simplified for practical application under longer-term exposure conditions. Field data is then used to estimate the variability in maximum pit depth at a given exposure location, derived from the application of extreme value statistics. Some comments are made also about the application of the simplified model.
Calcareous deposits on and within corrosion products tend to inhibit the (abiotic) corrosion of steels in seawater. Herein, it was considered whether this inhibition effect extends to microbiologically influenced corrosion (MIC) for extended (long-term) exposure periods. Quantitative estimates of corrosion rates were made from reported observations for 46 iron and steel shipwrecks, and other iron and steel objects immersed in seawater at various depths and for extended periods (many around 60 years and some up to 160 years). The observations are correlated with observations of the occurrence of calcareous deposits and information about dissolved inorganic nitrogen (DIN), a critical micronutrient for MIC. The results show that calcareous deposits can inhibit both long-term abiotic corrosion and long-term corrosion influenced by conditions suitable for MIC. The practical implications are briefly reviewed.
The present paper describes an interpretation of the variation in material loss of aluminium specimens in sea water immersion at 14 different locations worldwide. Careful examination of the conditions under which aluminium corrodes in very low velocity sea water suggests that the only significant variables are temperature and aeration. This allows the field observations obtained as part of an ASTM sponsored worldwide 'round robin' test programme to be viewed largely as a function of temperature. It is found that corrosion tends to be generally low except at temperatures around about 10°C and in the region 20–27°C with a local peak at around 23°C. This pattern applies for all exposures to 5 years.
A transcript of a discussion on the non-structural engineering aspects of buildings. The discussion also focuses on the special needs which are important for the construction and maintenance of heritage buildings.
Life extension of ageing steel structures such as naval ships requires consideration of plate thickness losses due to corrosion, particularly when protective measures such as paint coatings and sacrificial protection are not entirely effective. Traditionally, corrosion prediction models for ships take no account of the operational profile. Consequently, the corrosion models have an inherently high variability with poor corrosion prediction capability. A new corrosion model for the prediction of corrosion loss in the seawater ballast tanks of naval vessels has been developed. The model incorporates previously available corrosion models for immersion corrosion and atmospheric corrosion and takes account of operational and environmental variables. Experimental validation is presented for a trial on an operational naval vessel.
The corrosion process in large-particle porous media is investigated with particular reference to corrosion in bulk carriers (ships). Examples of large-particle porous media include raw materials such as coal and iron ore which typically have particle sizes ranging up to approximately 50 mm diameter. It is proposed that the corrosion process is dominated by finer particles which co-exist with the larger particles. A mathematical model of the corrosion process is developed and used to estimate corrosion rates. These are shown to compare well with experimental results. The analysis indicates that to minimise corrosion it is desirable to maintain the moisture content of porous media in contact with mild steel either to less than 50%, or to more than 80%, of the maximum water holding capacity of the media.
No abstract is provided for this article.
Theoretical and laboratory studies indicate that the corrosion of mild and low-alloy steels should be a function of temperature. In practice, however, it is usually held that this cannot be discerned from actual field corrosion data, being confounded by many other influences. The present study considers data from a recent ASTM-sponsored field study, which reported observations for more than 5 years for the immersion corrosion at 14 different sites around the world for copper-bearing carbon steel (UNS K01501). It also considers data reported historically in the literature, both for copper-bearing steel and for structural-grade steel. For these steels, general or uniform corrosion is the main interest because pitting is usually limited. When data is restricted to conditions that might be expected to prevail at sea and is carefully scrutinized and supplemented by information about (the likely) environmental conditions, it can be arranged to suggest that there is a clearly defined effect of temperature on corrosion behavior. This effect is consistent with laboratory observations but extends them in the lower temperature region even though the available data in this region is sparse. The dependence on temperature for steel is not inconsistent with earlier reported observations on the temperature dependence of 90:10 copper-nickel and aluminum. The effects of the degree of dissolved oxygen saturation and dilution by fresh water (and calcium carbonates) are apparent in the reported data, as is the influence of water velocity.
The conventional procedure for calculating the strength of reinforced concrete frames consists of an elastic analysis with, sometimes, a limited amount of moment adjustment permitted to allow for inelastic effects. From such an adjusted moment field, the cross-sectrional properties may then be checked by so-called ultimate load principles. It is well recongised that this procedure is quite irrational, and numerous attempts have been made to provide a more consistent method for non-linear reinforced concrete frame analysis. This paper reviews those methods briefly, and compares the predictions and foundations of these methods to experimental observations on real or model frames. It will be shown that the theories singularly ignore the behaviour of beam-column joints, which are shown to play an important part in real frame behaviour. The number of the covering absatract for the conference is TRIS no 394142. (Author/TRRL)
Steel pipes used in the oil and gas industry when subject to corrosion pitting may fail as a result of internal pressure and metal cracking influenced by hydrogen embrittlement. The progressive shape of the crack, including at the instant of failure, is predicted under increasing internal pressure. It is proposed that when hydrogen-assisted cracking (HAC) is involved the shape of the crack front may be estimated from a parallel numerical analyses in the absence of HAC. The theory for this is outlined and an example given.
For structural systems of practical size, it has been recently recognised that identification of all the stochastically dominant failure modes is a key to further development in reliability theory. This paper presents a search algorithm called 'truncated enumeration method', which generates only stochastically dominant modes and satisfies a specified truncation criterion. An example is demonstrated. The relationship of this method to other techniques is noted (a).
No abstract is provided for this article.
Pitting corrosion of aluminium alloy AA5005-H34 immersed in natural sea water for up to 2 years is presented, together with new interpretations of the development of the pitting process over extended exposure periods. Trenching initiated at surface-located Fe-rich intermetallic (IM) particles. It occurred more extensively in sea water compared to atmospheric exposures. With increasing exposures, the pitting morphologies became more complex. Three types, i.e. hemispherical pits (Type 1), crystallographic pits (Type 2) and characteristic 'petal'-like pits (Type 3), were observed after 12 months. It is suggested that the formation of Type 2 and Type 3 pits occurs at an advanced corrosion stage, and are associated with large Fe-rich IM particles near the metal surface. The influence of marine growth on pitting was negligible.
This paper reviews recent developments in the prediction of the likely future corrosion losses and of the maximum pit depth for steels exposed to marine environments. A robust mathematical model based on corrosion science principles and calibrated for immersion conditions to an extensive range of literature and new data is described. The model has provided explanations for the effects of steel composition, water velocity, depth of immersion and seawater salinity and also has facilitated new interpretations of data for long-term pitting corrosion. This paper briefly overviews these developments and refers to some typical applications, including marine corrosion of ship ballast tanks, corrosion of sheet piling in harbours and corrosion of offshore platform mooring chains.
Reliability theory for the assessment of existing structures for remaining safe life must consider the probabilistic deterioration of structural strength as well as multiple applied loadings. This requires good-quality models of deterioration processes. Models for corrosion, which have been applied in the literature for marine applications, are reviewed. It is shown that these usually are statistical only, using data pooled from many sources. They also use little or no theoretical insight. As a result, they provide poor-quality mean-value information and very high statistical uncertainties. The conclusion is that better modeling is required and that this involves much better understanding of the factors influencing marine corrosion. These factors are reviewed briefly, with particular emphasis on marine immersion corrosion of mild steel.
