
You could do worse than pack a punch with a hard hitting quotation to begin an article with. So here it is: “Safety is not an intellectual exercise to keep us in work. It is a matter of life and death.
It is the sum of our contributions to safety management that determines whether the people we work with live or die.” This is what Brian Appleton, Technical Assessor to the Piper Alpha inquiry said – re-quoting it from Trevor Kletz’s publication, Lessons from Disaster, 1993.
Regrettably, over the years there have been too many globally significant incidents associated with the petrochemical industry in which detection and/or fire protection control have been lost. This almost certainly leads to fire and/or explosion – and loss of life in almost all cases.
Each has presented an opportunity to learn and improve risk control management. Flixborough in June 1974, Piper Alpha in July 1988, the Buncefield Storage Tank farm fire 2005, Texas City Refinery explosion in 2005 and the Deepwater Horizon offshore rig fire and explosion in 2010 are just some examples.
There is generally a more heavily enforced and complex legal and regulatory framework applied to the petrochemical industry outside of the Middle East. Yet despite their influence and the potential consequences of not adhering to regulations, these catastrophic incidents of global significance continue to occur outside of the Middle East, yet apparently not within it.
The lengthy and expensive investigations that are conducted almost always bring about new or amended laws, regulations, codes of practise and guidelines. All the same, repetitions of the same failures of control recur – history is witness to this. These recurrences might be seen as an indictment of the petrochemical industry’s unwillingness to learn, or to implement the necessary changes to operations.
Most large petrochemical companies in the Middle East do their best to adopt the occupational safety and health standards that are imposed by law, whether international or national, upon the large global players – Shell, Occidental, BP, Total and so on. At least, they integrate them within their impressive HSE Management Systems. Reliable implementation by staff in the field, however, particularly contractor staff, remains a challenge.
The UK has multiple pieces of law that are applicable, including the European Union’s Seveso II and more recently, III Directives, the Control of Major Accident Hazard Regulations (COMAH), Dangerous Substances Emergency Action Regulations (DSEAR), the Regulatory Reform (Fire Safety) Order (RRO), and numerous others that have a bearing on the safe operation of petrochemical plants.
Best Practise guidance is issued by industry professional organisations such as the Energy Institute (EI) and the American Petroleum Institute (API). The International Labour Organization (ILO), part of the United Nations organisation, has a range of conventions, treaties and guidelines related to the petrochemicals industry that most signatory nations have signed or adopted. Many of these deal with the manner in which hazardous materials should be handled in order to protect the occupational safety of workers. What the majority now seem to have in common is a less prescriptive approach and an advancement of the principle of risk based solutions.
Petrochemical facilities take various forms. Generally speaking there are simpler facilities in the upstream operation – collector stations, production and storage facilities – and more complex downstream plants where raw products are stored in larger quantities, processed in one or more ways and the resultant product(s) distributed.
Add to this the complexity of materials logistics, distribution terminals and offshore facilities connected to onshore plant and, without appropriate controls and emergency planning, the potential for an incident involving a hazardous substance to occur is, in relative terms, high.
With a multiplicity of organic and inorganic hazardous materials that have very low flashpoints, low LELs and a wide flammable/explosive range found within the industry, fire and explosion represent two of the primary hazards to be controlled, pretty much at every stage in the production chain. The need to handle or process these hazardous materials presents additional potential risks to staff operating the plants.
Petrochemical plant lifecycle phases
Excluding the design phase, the life of a facility or plant is covered by five activity phases: construction, commissioning, operation, decommissioning, and demolition. What form should the control of fire and explosion hazards take in each of these phases?
It is essential to be able to identify a hazard-driven threat to life, the environment and assets as early as possible, raise the alarm and enable executive actions to be taken to prevent the risk being realised.
As the lifecycle phases change, so do the nature of the hazards and the degree of risk they present. The solution selected must reflect the outcome of the review of potential credible scenarios and subsequent risk assessments.
The most challenging phases will be those where major change is occurring related to the hazards and risks, i.e. construction and demolition. Next to these would be the commissioning and decommissioning phases. In terms of control management, control of the operational phase ought to be the easiest to achieve, yet the recorded catastrophic incidents all seem to occur during this lifecycle phase. Some explanation may come from the realisation that in each case, ‘standard’ operating parameters were not being followed; that is, change occurred, it was not managed or controlled effectively, and the incident occurred.
During construction, commissioning, decommissioning and demolition, the designed operational phase detection and fire protection systems will be incomplete for much of the time. Any associated control systems that determine the executive actions that need to be taken when alarms are raised will be in a similar state. Many of the flammable/explosive/toxic hazards may still be present on site, although the inventory is likely to be markedly reduced.
Prior to any of these lifecycle phases commencing, alternative arrangements for detection and fire protection need to have been considered against the potential credible scenarios, and the selected solutions implemented. Permit to Work systems, with associated isolations, certifications and gas testing arrangements will only go so far.
At these higher risk stages, it would also be considered essential to ensure that all staff engaged in the planned activities have been trained in the detection and alarm arrangements, and particularly the fire protection arrangements. It is highly probable that for much of the time, active fire protection measures will rely on human intervention. To be effective, the workforce must be thoroughly trained and prepared.
In the Middle East particularly, there are many petrochemical facilities established in remote areas where Civil Defence or Industrial Fire and Rescue Services are not available at all, or only available after considerable delay due to travel distance and terrain.
Hierarchy of controls
The widely accepted principle of the hierarchy of hazard controls does provide some guidance on how to deal with the fire and explosion hazards issue. The higher in the list a control is, the more effective control it is likely to provide.
Elimination of the material being used or processed, at the top of the list, is probably the least likely control option that the petrochemical plant operator has. Substitution of a hazardous material to one with a higher flashpoint, LEL, or less toxicity may be possible in some cases, but in the main, engineering controls are most commonly going to provide the highest degree of control that is practicable.
At the plant and process design stage, the HAZOP study reduces the risk of undesirable and unexpected events being realised. Coupled with Failure Modes and Effects Analysis, Fault Tree Analysis and Event Tree analysis, designers ought to be able to design out undesirable outcomes, and design in additional controls where there is an identified ‘weakness’, that at least become ‘live’ once the facility has been commissioned.
Engineering controls
Detection of a desired or undesired event or condition in a plant or facility which enables an action to be taken, is an engineering control. Fire protection, whether passive or active, is also an engineering control.
In simple terms, the objective of a detection system is to reliably detect a desired or undesired event at the earliest possible moment – and indicate the detection has occurred.
Again, in simple terms, the objective of fire protection is to protect equipment and structures from damage caused by any form of heat produced by prolonged exposure to fire and/or flame.
What is it we are seeking to protect and keep safe? Remember the PEAR acronym: People, the Environment, the Assets, and finally Reputation. When combined, detection and fire protection have the ability to ensure the means of escape inside buildings on site, and from the plant structures themselves, are always available and not compromised. In addition they can provide prevention, protection and mitigation of the effects of fire on plant and processes. This means that both detection and fire protection can be used for life safety purposes, as well as others.
In relation to the prevention or control of fire, detection may be installed to identify:
• The presence of a condition that if not addressed will lead to loss of primary containment of a flammable, explosive and/or toxic substance
• The presence of a flammable, explosive or toxic substance in an area where it is not wanted
• The presence of a hazardous substance in a condition that may lead to fire and/or explosion if not addressed
• The presence of fire, heat, smoke or combustion products
• A threat from any of the above that would compromise means of escape for persons at work
When integrated with control systems, activate executive commands to:
• Prevent event escalation
• Reduce the effects of the event on critical components
In relation to the continued safe operation of the plant and its processes, fire protection offers:
• A passive resistance to the effects of heat from fire on structural and other critical components of the plant
• An active response to fire, or the conditions that may lead to it breaking out, which will:
Prevent ignition from occurring Protect surrounding critical risks from the effects of heat from the fire Attack the fire to reduce its ability to escalate Extinguish the fire Reduce or prevent any potential re-ignition
Detection
Detectors linked to an alarm system and integrated into control systems provide a Safety Instrumented Function (SIF), which is capable of providing some Automated Executive actions, e.g. switching things on or off, closure of fire dampers, AC shutdown, positive air pressurisation of escape staircases, starting total flood systems, drenchers, water spray projection or foam systems. All are engineering controls.
There is a wide range of detectors available. Selecting the right kind of detection to adequately deal with each lifecycle phase will be quite a challenge. In every case, selection is based upon the potential credible event scenario, and the nature of the hazard. For example, during construction or decommissioning, hot work is very likely. The nature and light frequency of a gas torch flame, a welding arc, or grinding sparks, however, are all very different from the light frequency emitted by a fire flame. This difference may assist the selection of the most appropriate detector type, for the credible potential scenarios.
Fire protection
Fire protection may be realised through passive and/or active controls.
Passive controls are those that either form part of the materials that make the structure itself, or materials applied to them. They remain in position and provide a barrier to heat and/or smoke transmission, e.g. a mild steel beam supporting a pipe rack in a process facility will lose two thirds of its structural strength when its temperature reaches 650° C. (A hydrocarbon fire can reach 900° C in eight minutes). After construction, if the steel is sprayed with a cementitious material (e.g. cement/vermiculite) to an appropriate thickness (BS 476 refers) up to six hours of fire resistance is provided.
In buildings, fire protection of escape routes are achieved by compartmentation (breaking a space up into fire resistant spaces) using fire resistant materials, fire-stopping where breaches occur for services access. Doors, walls, floors and ceilings are all considered, as are the surface treatments of the escape routes.
Active controls are systems that are in situ but provide no part in the prevention or protection until activated by a specific event, e.g. a sprinkler head plays no part in control or extinguishment until it is activated by heat at a specific temperature. An LPG sphere water drencher does not operate until a threat is detected and an executive command – automated or manual – is given to drench the tank with water to protect it from a heat threat.
Portable fire equipment, fixed fire equipment such as monitors, and semi-fixed such as trailer mounted monitors are all described as Active Fire Protection measures. All require an executive command in order to be brought to bear on the threat.
Communicating the lessons
Little is being heard about significant leaks, fires or explosions in petrochemical plants in the Middle East or the lessons that have been learned from them. Given the large number of exploration and production facilities across the Middle East region, an increasing number of petrochemical processing plants operating or under construction, massive distribution terminals for gas, oil and chemicals, and the generally hostile environment in terms of temperature and humidity levels, it is not surprising to learn of fires, leaks and explosions occurring from newspaper headlines and press agency reports. What is disappointing is the apparent absence of any detailed information that the global industry could learn from.
Statistically, there must have been similar large scale, significant incidents in the Middle East as have occurred in other parts of the world, but there are almost no records or reports of the lessons learned available in the public arena. Sharing loss experience in a transparent way so others may learn how to avoid a similar problem could be the gift of life to a worker in another part of the world.
References
Model Code of Practice, Part 19, Fire Precautions at Petroleum Refineries and Bulk Storage Installations, (3rd Edition), Energy Institute, 2012 Control of Major Accident Hazards (COMAH) 1999 Control of Substances Hazardous to Health (COSHH) 2002 ILO, C170, Chemicals Convention, Geneva, 1990 ILO, Fire Risk Management, Geneva, 2012 ILO, Convention 155 on Occupational Safety and Health standards, Geneva, 1981 ILO, Convention 170 on Chemicals, Geneva, 1990 ILO, Convention 174 on Prevention of Major Industrial Accidents, Geneva, 1993 ILO, Convention 187 on Promotional Framework for Occupational Safety & Health, Geneva, 2006 ILO, Final Report – Meeting of Experts to examine Instruments, Knowledge, Advocacy, Technical Cooperation and International Collaboration as Tools with a view to Developing a Policy Framework for Hazardous Substances, Geneva, 2007
Published: 26th Feb 2014 in Health and Safety Middle East