Corrosion Engineering,Science and Technology: The International Journal of Corrosion Processes and Corrosion Control

Further sessions of this international event held in Bournemouth in November 1994 are summarised in a report by A. D. MERCER. The report is compiled mainly on the basis of the published papers with the assistance of comments from the chairman or convenor of each session.

The inaugural Marcel Pourbaix Awards for Promotion of International Cooperation were presented at a ceremony organised by the International Corrosion Council (ICC) in Brussels in April. In another presentation Marcel Pourbaix was himself honoured in acknowledgement of his contribution to the international corrosion community.

This paper re-examines the reasons for the continuing high cost of corrosion failures and suggests that they are largely attributable to the influence of people on engineering systems. The paper also discusses the differences between corrosion control and corrosion management, the former being concerned with the interaction of materials and environments while the latter includes the additional effects of people. This systems engineering approach not only provides a clearer explanation of corrosion failures in engineering, but suggests a strategy for improved systems performance and cost effective engineering in the future. The way ahead, through the use of machine based intelligence, should help to make decisions more reliable and reduce the recurrent negative impact humans often have on systems, either due to negligence or more commonly to ignorance. The possibilities for the development and application of such computer systems are discussed.

The fatigue crack propagation (FCP) behaviour of peak aged Al—Zn—Mg—Mn alloy has been investigated with reference to that of peak aged Al—Zn—Mg alloy by comparing crack closure in 0·5M Na₂SO₄ + 0·5M NaCl and 0·5M Na2SO4 solutions. Anodic studies were complemented by potentiostatic current transient measurements. For both alloys, the intrinsic FCP rate in the Cl^- containing solution exceeded that in dry air; however, the apparent FCP rates were similar. Crack closure decreased in the order: 0·5M Na₂SO₄ + 0·5M NaCl solution, 0·5M Na₂SO₄ solution, dry air. Irrespective of environmental conditions, the addition of manganese to the Al—Zn—Mg alloy reduced the intrinsic FCP rate and crack closure. Specimens in the Cl^- containing solution were more slowly repassivated than those in the sulphate solution. In the Cl^- containing solution the Al—Zn—Mg alloy was more slowly repassivated than the Al—Zn—Mg—Mn alloy, in good agreement with the trend of intrinsic FCP. The enhanced intrinsic FCP rate and crack closure in the Cl^- containing solution resulted from an increase in localised corrosion attack on the slip step at the crack tip.

The exfoliation corrosion behaviour of sheet and plate materials of various conventional aluminium and Al–Li alloys has been evaluated using accelerated tests. Results are .compared with atmospheric exposure data published in the literature to assess the applicability of the testing techniques employed. For damage tolerant Al–Li based sheet and plate, the cyclic acidified salt fog (Mastmaasis) test according to ASTM G85, Annex A2 indicated susceptibility to exfoliation corrosion, reproducing the limited outdoor corrosion data for the Al–Li alloys 8090–T81 and 2091–T84 as well as marine exposure results reported for the conventional alloys 2024–T351 and 7075–T7351. Therefore, it appears to be a promising testing technique for predicting the service performance of high strength aluminium alloys. Compared with the ratings determined following the cyclic acidified saltfog tests, the standard Exco test according to ASTM G34 indicated better exfoliation corrosion behaviour of the alloys investigated, except for 8090–T6 sheet and 7075–T7351 plate, which exhibited severe and mild exfoliation respectively. In the modified Exco test suggested by Lee and Lifka, 7075–T7351 panels were susceptible to pitting, whereas the other alloys studied generally suffered more severe exfoliation than in the standard Exco test.

The construction and operation of an electrochemical sensor capable of monitoring the corrosion performance of zinc alloy coated steel under controlled atmospheric conditions are described. Sensors were fabricated from zinc and zinc alloy coated steel sheet with and without chromate conversion treatments. Corrosion behaviour was monitored by measuring the potential and polarisation resistance of the coating and the galvanic current flowing between the coating and a gold cathode. It has been shown that this sensor is sufficiently sensitive to detect different corrosion rates for different types of zinc alloy coating in a similar environment, changes in the corrosion rate associated with changes in environmental conditions, and increases in corrosion activity associated with defects in the coatings. The sensor has potential for development as a simple and inexpensive method of assessing relative long term performance of a wide range of zinc alloy and other metallic coating systems, in a range of environments.

Sensors whiCh use electrochemical impedance spectroscopy (EIS) to monitor in situ the corrosion of polymer coated metals in atmospheric conditions have recently been developed. The validity of these sensors, which consist of a gold grid deposited on top of the organic coating, has been tested in a cyclic corrosion test in which electrocoated samples sprayed with 0·5 or 5·0 wt-%NaCl were exposed to wet and dry conditions for alternate periods. It was found that when delamination was significant, the equivalent circuit should include the interfacial impedance of the gold electrode. Furthermore, it was observed that the features of the EIS spectra were dependent on both the delamination rate and the conductivity of the thin electrolyte film present on the paint during exposure.

The photocorrosion of n-GaAs electrodes in aqueous solutions represents one of the most important factors limiting their application in solar energy conversion. In continuation of previous studies, the corrosion of n-GaAs has been investigated in 0·5M H₂SO₄ using a pllOtoelectrochemical cell. Illumination of the n-GaAs anode with white light of intensity 20 mW cm^-2 generated a saturation photocurrent of 27 mA cm^-2 at biasing voltages more positive than +0·45 V(SCE). This current corresponded to hydrogen discharge at the platinum counter electrode and to oxidation of the GaAs surface. When 0·01-6·0M LiI was added to the H₂SO₄ solution, 12 was formed at the n-GaAs electrode and dissolved in the iodide solution. The onset potential of the photocurrent (E_p) shifted from -0·36 to -0·73 V(SCE) as the concentration was varied from 0·01 to 6·0M I^-. Addition of I^- ₃ shifted E_p to more positive values. The extent of corrosion was studied by comparing the ratios of photogenerated I₂ and H₂. In 0·01, 1·0, and 6·0M I^- solutions, I₂/H₂ was 0·38, 0·87, and ∼1 respectively. These ratios imply that, at lower I^- concentrations, the holes photogenerated at the n-GaAs surface were consumed in the simultaneous oxidation of I^- and GaAs. However, in 6·0M I^-, where I₂/H₂ = 1, surface oxidation of GaAs was completely suppressed owing to total consumption of the photogenerated holes by the oxidation of I^-. The results are considered on the basis of the band model of the semiconductor/electrolyte interface.

In order to test the applicability of 8090-T81 alloy in a marine environment, its corrosion behaviour was investigated in natural sea water at 25°C. Corrosion resistance was examined using free corrosion and electrochemical tests. The swface film was characterised by ir spectroscopy, X-ray microdijfractometry, and X-ray photoelectron spectroscopy. Microstructural heterogeneity increases susceptibility to localised attack, particularly to crevice corrosion, thus discouraging any marine use of this alloy in a non-protected form.

The corrosion susceptibility of untreated and anodised aluminium alloy 1050, with and without the presence of corrosion products, has been investigated in 3·5%NaCl solution and simu.lated sea water. Corrosion rates in the presence of corrosion products were calculated from weight loss measurements under an applied anodic current. This procedure ensured the development and accumulation of corrosion products in the solution and on the alloy surface. Experiments in the absence of corrosion products were carried out under potentiodynamic polarisation at a high scan rate and corrosion rates were calculated from polarisation curves. Anodised coatings up to 20 μm in thickness were found to protect the alloy in 3·5%NaCl solution under both sets of experimental conditions. In sea water these coatings protected the alloy only in the absence of corrosion products. In the presence of corrosion products, higher corrosion rates were observed in sea water than in 3·5%NaCl. Higher corrosion rates and anodic currents were also observed in sea water in the absence of corrosion products, but the corrosion and pitting potentials moved to more positive values. An explanation is given of the results, based on the protective properties of anodic coatings, the influence of sea water composition, and the influence of corrosion products.

Tests were carried out to investigate the ability ofZeron 100 super duplex (UNS S32760) and 6Mo superaustenitic (UNS S31254) stainless steels to repassivate during short temperature upsets in sea water cooling systems (70°C). The results showed that over 50% of the. Zeron 100 specimens failed to initiate crevice corrosion, and repassivation of the other specimens occurred quickly during cooling. Tests on 6Mo austenitic stainless steel showed that the specimens readily suffered crevice corrosion, and repassivation only occurred at about 25 to 30°C. The results are compared with those of other workers with other high alloy stainless steels. Service experience supports the laboratory results and shows the high resistance of Zeron 100 to crevice corrosion in chlorinated sea water.

Alternating voltage corrosion of types 301, 304, and 316 stainless steels in an aqueous boiling NaCl solution under continuous reflux was investigated through weight loss measurements as a function of time. Specimens of each type of stainless steel were used as electrodes to heat the solution by the Joule effect using pure outlet alternating voltage modulation (127 V, 60 Hz). Pitting corrosion was observed in all cases, but it was severe for 304 and less marked for 301. This led to average percentage weight losses, after 1500 h of experiment, of up to 30%, 21%, and 16% for the 304, 316, and 301 steels respectively. The enhanced pitting corrosion of the 304 steel agrees with results previously reported in the literature by other authors.