ICORR & GUESTS:

Oladimeji Apesin – Andrew Palmer & Associates

Bill Deans – Kings College, Aberdeen

Chris Googan – Anti Corrosion Engineering

Dan Kirkwood – Oceaneering International

Andrew Nowicki – PSN

Bopinder Phull – Corrosion Consultant, ex LaQue

Ernesto Santana-Diaz – DNV

Brian Struszczak - DNV

Ed Wade – MetalEcosse

APOLOGIES  

1.  TECHNICAL PRESENTATIONS 

Click for abstracts 

1.1  'Some Marine Coating and CP Failures', Chris Googan, Anticorrosion Engineering

1.2  'Electrochemical Behaviour of CuNi 90/10 in Chlorinated Seawater under Stagnating Conditions’, Wilhelm Schleich , KM Europa Metal

2.  OPEN FORUM

Dan Kirkwood (Oceaneering International), by request, gave details of an investigation with which he had recently been involved.  The submission was based on a fairly recent investigation into the early appearance of blisters in a glass flake vinyl ester internal coating of an oil and gas separator vessel after a very short period of service (approx 1- 2 years).   Potential litigation and business insurance issues required that an extensive and thorough investigation be carried out on the blistering mechanism involving analysis of samples of the service coating including autoclave tests at different temperatures, SEM, DSC and DMTA work on the coating materials, EDAX work on corrosion products and chemical analyses of fluids from different sizes (maturities) of coating blisters.  

There were a number of interesting outcomes from this work one of which was the conclusion on the root cause of the blisters.  All the high integrity blisters examined (i.e. those where the coating cover had been uplifted but were still intact with no gross fluid pathways microscopically evident) were completely filled by aqueous fluid with varying solute salt concentrations.  It was found that sodium and chloride ions were the predominant species present.  Some other significant patterns were found.  Generally the larger blister fluids had lower salt concentrations than the smaller blister fluids.  A model was produced showing that osmotic diffusion of water from the separator cargo fluids had taken place through the coating acting as a semi permeable membrane and driven by residual salts left on the steel surface prior to coating application.  Close estimates of steel surface areas within blisters and salt concentrations found in those associated blister fluids led to the conclusion that, on average, around 25 - 35 mg/m² of soluble salts appeared to have been left on the steel surface prior to coating.  Had these salts been absent from the steel surface when it was coated there would have been no driver for osmotic diffusion and the particular blister appearances.  

One message was clear.  It has been recognised for many years that residual salts on steel surfaces were undesirable and there have been general and welcome moves to reduce the prescribed tolerable levels of residual salts particularly in the past 40 years or so.  NASA and US Navy studies in the 1960s and 1970s were first to recognise the importance of analysing and prescribing levels of residual salts as part of a full surface preparation prescription.  70 mg/m² appeared in early papers in the 1960s and early 1970s, though this `tolerable level` has steadily been driven down to a maximum of 20 mg/m² (as specified in NORSOK M 501, June 2004 for example).  Depending on how this salt is disposed on a steel surface (i.e. salt `clusters` are more likely than even salt distribution over a steel surface) this figure is probably still too high to prevent osmotic blistering of coatings in aggressive environments.  Not only that, there remain considerable challenges in effectively sampling and measuring residual salts on steel surfaces. 

There can be little doubt now that many premature coating defects and failures reported over the years have actually had a root cause in osmotic related mechanisms including cold wall effects driven by residual salts on metal surfaces, rather than latent defects in the coating materials themselves.  Even the present `tolerable level` is likely to be excessive for high performance coatings to cope with over a reasonable expected lifetime, certainly in demanding corrosivity conditions. Ultimately the only acceptable figure for residual salt must be 0 mg/m², but realistically this is bound to be near impossible to both achieve and measure.

3.  TECHNICAL PRESENTATIONS (part 2)

3.1  ‘The Effect of Surface Finish on Corrosion Behaviour of Super-duplex Stainless Steel',  Bill Deans, Department of Engineering, Kings College, Aberdeen