News - full story |
|
22/9/2009 Ignorance blamed for fatal explosionCSB calls for mandatory reactive chemistry education |
|
T2 blast was "one of the most destructive ever investigated" |
|
PLAIN ignorance on behalf of the chemist and chemical engineer running a small family-owned chemicals plant has been blamed for a runaway reaction that caused the death of four people – a third of the plant’s workforce. The US Chemical Safety and Hazard Investigation Board (CSB) called the 19 December 2007 accident “one of the most destructive ever investigated by the board.” The explosion was caused by a runaway chemical reaction that likely resulted from an inadequate reactor cooling system, the investigators say in their final draft report. The accident occurred during a production run of the fuel additive MCMT (methylcyclopentadienyl manganese tricarbonyl), which T2 produced using a 9500 l batch reactor. The CSB investigation found that the recipe used by T2 created two exothermic reactions; the first was an intended part of producing MCMT but the second, undesired reaction occurred if the temperature went above 390ºF, slightly higher than the normal production temperature. On the day of the accident, CSB says that the cooling system was not working, probably because of a blocked pipe or a failed valve. The ensuing a runaway reaction caused the temperature and pressure inside the reactor to rise uncontrollably. Ten minutes later, it exploded with the power of 1400 lb of TNT. Four T2 employees were killed and 32 people were injured, including 28 members of the public. CSB says it found reactor debris a mile away from the reactor a mile away from the site. Safety experts contacted by tce all agree that the accident was preventable and that a proper HAZOP would have pinpointed the inherent dangers of the reaction. Robert Hall, who supervised CSB’s investigation, says that T2 did not appear to sufficiently understand its own processes and its inherent dangers. “Despite a number of near-misses during earlier production efforts, T2 failed to recognise the underlying runaway reaction hazard associated with its manufacturing process,” he says. Robin Turney, chair of IChemE’s Loss Prevention Panel and an independent safety consultant, adds: “Clearly this was a very hazardous reaction. Whilst the full details of the Turney says that appropriate control measures which should have been considered would include the use of a ‘short-stop’ additive to halt the reaction or the provision of a relief system capable of handling the material generated by the runaway reaction. “An intrinsically safe approach would involve studying whether the reaction could be carried out in a small continuous reactor or by adding one of the reagents steadily during the reaction in what’s known as a semibatch process, thus minimising the unreacted material present within the reactor.” However, it is not clear whether T2 had considered taking this approach. The CSB report notes that there was no design redundancy in the process, making it susceptible to a single-point failure, and that the reactor relief system was not able to relieve the pressure built up by the runaway reaction. Its main recommendation is that all universities should teach reactive chemical engineering – currently less than 11% of US universities include it in undergraduate chemeng degrees. Experts however are sceptical whether undergraduate courses could provide enough detail, and that the focus should be on continued professional development. David Shallcross, IChemE’s vp/education, says: “Graduates would be expected to learn more of the actual safety procedures and process controls with the company they work for, through either company training or continuing education. The purpose of undergraduate training should be to provide the graduate with sufficient background to allow them to learn more about the specific cases that they find in their employment. Graduates must – and most do! – understand the importance of life-long learning.” James Bickerton, health and safety advisor for the consultancy First Business Support, points out batch and semi-batch reactions are known to be hazardous. The risks of these reactions were understood in the 1950s, when ICI Dyestuffs set up a protocol to assess all chemical reactions carried out on greater than research-lab-scale for their chemical reaction risk, he says. Despite this, accidents caused by runaway chemical continue to occur. For example, the 1976 Seveso disaster, where a cloud of sodium hydroxide, toluene and dioxin was released from a plant operated by the Italian company ICMESA, leading to widespread contamination, was the result of a runaway reaction. Other examples include Hoechst’s 1993 o-nitoanisole leak at Griesheim, Germany; and the 1996 explosion at Albright and Wilson’s Avonmouth site in the UK. “It seems that the lessons from the past have not been learnt or haven't crossed the Atlantic,” says Mark Hailwood, a scientist working for the German safety regulator LUBW. The underlying view is that chemical engineers involved in designing or operating plants should take responsibility for their processes and make sure they have all the relevant knowledge to understand and control all the reactions that could feasibly occur at their site. “The main lead has come via IChemE – events such as the Hazards Symposia and the Loss Prevention Bulletin, which has been set up specifically to share the lessons learned from past accidents,” says Bickerton. “There is a lot of information available to anyone who is aware of the potential problems.” Reactions hazards is also one of the topics covered in the new IChemE course Fundamentals of Process safety which is aimed at providing engineers, managers and supervisors with an understanding of how accidents happen and the importance of understanding and following best practice. |
|