Taking a closer look
The number of incidents that have happened as a result of tank corrosion have diminished significantly over the past few years. But when a tank failure does occur, it can be disastrous, so regular tank inspection is vital.
Since April 1999 when the Control of Major Accident Hazards (COMAH) regulations came into force in England, Wales and Scotland, the Competant Authority has been required to notify the European Commission of certain major events. As a consequence, there have only been two incidents involving storage tanks affected by corrosion in the UK - between the onset of the regulation and the end of 2005.
Tank failures
In July 1999, Esso Petroleum at Fawley, UK, experienced a failure in a tank bottom resulting in the release of 400 tonnes of crude oil, which was contained within the bund. Esso amended its tank inspection programme to the satisfaction of the Health and Safety Executive (HSE). Just two months later, the Vopak storage terminal at Middlesbrough, UK, suffered a 12 tonne spillage of sodium cyanide solution due to corrosion of a defective weld. The solution soaked into the ground and ran into the River Tees, resulting in Vopak receiving a prohibition notice, a £5,000 (€7,412) fine for polluting the river, and a repair bill of £245,000.
Despite the improved history in the UK, in November last year the HSE issued a safety alert relating to the rupture of an atmospheric crude oil storage tank that occurred in Belgium in October 2005. The rupture affected a 40,000 cubic metre tank that was more than three-quarters full, with the leak being noticed in the refinery control-room through the activation of the low level alarm. The storage tank was built in 1971, had been inspected in 1990 and put back into service by a new owner in 1991. An external inspection was undertaken in 1994, and then subsequently every three years. A full inspection was scheduled for 2006, although in 2004 measurements were performed that showed no abnormal features.
Investigations determined that a gutter had formed in the bottom of the tank 1.5 metres from the tank shell, and that the accumulation of stagnant water had caused internal corrosion that reduced the thickness of the bottom plates in that location. No other plates in the area displayed indications of severe corrosion.
It is thought that the gutter formed due to settlement in the tank foundation, which consisted of compacted sand. The gutter in the bottom plates had not been detected during the internal inspection in 1990-1991 most likely due to the technique used, and also because, with the tank empty, elastic deformation would have partially hidden the gutter. The visual inspection found pitting corrosion, and ultrasonic thickness measurements were performed on the affected plates that were subsequently repaired.
All the other tanks on the site were examined and found to have the same defect located in the same position. Although the ability to see the defect was variable, in each case the thickness of the plate in the gutter was reduced and in some cases there were small perforations. A small leak in one of the other tanks was shown to have happened a few days before the major tank rupture, but the foundations had moved sufficiently to seal it.
Whether the lack of
incidents is the result of
more effective inspection
programmes, improved
techniques, better operator
application or just good
fortune, will probably not be
known for another 10 years
or so. What is obvious,
however, is that the results
of a tank failure can be so
catastrophic that research
into inspection methods is
continuing and new
technology is reaching the
market.
