Tank fire engineering: the big picture

Storage facilities, firefighting foam, and the assessment of risk, by Roger A Klein, Cambridge, UK.

Published:  05 April, 2011

Storage facilities holding large quantities of hydrocarbon fuel or polar solvents for the chemical industry, sea and airports represent some of the major potential users of fire fighting foam. Unlike municipal fire services and aviation rescue and fire fighting (ARFF) incidents, storage sites can be adequately bunded so that fire water run-off would be contained at an incident thus minimising the risk to the aquatic environment.

In purely practical terms, adequately bunded must be taken to mean that the bund volume – bund area times bund-wall height – should be equal to or exceed the volume of the contents of the storage tank(s), plus that of any fire fighting foam used, plus that of cooling water needed to prevent structural steel damage to the tank itself or surrounding tanks. Bund design and tank layout are critical in reducing risk to the aquatic environment from contaminated run-off.

 

Any large storage site is required under local legislation, such as the UK Control of Major Accidents and Hazards (COMAH) Regulations, to carry out some form of risk assessment and to have a risk management or major incident management strategy.

The elements of a suitable and sufficient assessment of risk have been outlined in the UK HSE’s document HSG65, first published in 1991 with a second edition in 1997. The elements of any risk assessment include (i) identifying the physical presence of a hazard, for example, 100,000 litres of ethanol; (ii) identify the hazard’s capacity to do harm, ie, the risk – this involves identifying the target(s) at risk whether this be a particular sector of the population, the environment, plant facilities and infrastructure, legal and financial implications, the organisation image and brand name, or a wider societal impact; (iii) identifying measures to either (a) eliminate the risk, (b) reduce its consequences, or (c) if neither of these are possible under the circumstances, eg, at an emergency incident, provision of suitable protective personal equipment (PPE) or other containment measures; (iv) real-time monitoring procedures which might include measurement of toxic gas levels, for example; and (v) robust auditing procedures to ensure that the system of risk management works!

 

Over-arching these general principles the organisation must have a clearly defined, written risk management policy, supported by an infrastructure that is fit-for-purpose as well as by management at all levels including those responsible for financing the system. Non-compliance with the organisation’s risk management policy by managers at any level however senior should be considered a disciplinary matter with the appropriate sanctions applied.

 

Having identified the hazards present and the risks these pose, one of the key policy or strategic issues is the choice of appropriate fire fighting measures taking into account all the risks present. These may include tank fires, ruptured tank and bund fires, running pool fires, boiling liquid vapour explosions (BLEVEs), and vapour cloud or gas fires.

Site design should provide adequate bund volumes and adequate tank spacing – unfortunately not always the case – containment for excess run-off such as an isolatable foul water system, remotely operated shutoff valves (ROVs) to close down process lines, as well as flame-proof pumps and switchgear.

 

One of the problems at the Flixborough incident in 1974 was that the diesel-operated pumps continued to function even when apparently shut down because the surrounding air was so rich in cyclohexane that this air-fuel mixture was drawn in through the pumps’ air intakes with the result that the engines continued to compression fire (‘dieseling’).

 

It is also essential that foam stocks are kept well separated from the potential incident site or they may be lost if the incident is large enough – this happened at the Buncefield incident in 2005. Foam inlets for fixed installations should be placed far enough away from large tanks that firefighters will be able to use them in case of a large bund fire.

 

Risk assessment and planning must take account of potentially very high levels of radiant heat output from a large bund or running pool fire; moreover, the risk of significant blast over-pressure affecting surrounding structures and equipment should be allowed for.

 

Radiant heat levels together with fuel burn-off rates and probable flame height can be estimated using a variety of commercially available software. Characteristic pool-fire burning rates in kg/sec/m2 have been reported by Babrauskas[1]  for some common fuels and flammable organic solvents such as gasoline, kerosene, JP-4 and JP-5, heptane, xylene, ethanol and dioxane.

 

Combustion energies for a wide range of fuels are also available and are measured in megajoules per kilogram (MJ/kg). Oxygenated fuels such as ethanol, acetone or dioxane tend to have lower heat outputs in MJ/kg and lower burning rates than saturated hydrocarbon fuels.

 

Knowing the amount of fuel involved in kg (volume x density), the area of the pool or bund fire and the burning rate, it is possible to calculate the total theoretical heat output in megajoules per second, ie, in megawatts (MW). A simple calculation allows an estimate of how long the fire should burn unattended.

 

Taking as an example a 200 m2 bund containing 200,000 litres of gasoline (a layer 1 m thick), the fuel will burn off at the rate of 200 x 0.055 = 11 kg/sec and will produce a maximum heat output of 43.7 x 11 = 481 MW (4.807 x 108 joules per second). The 200,000 litres will take approximately (200,000 x 0.74)/(11 x 60) = 3.74 hours to burn-off. Heptane, on the other hand, would produce 901 MW and burn-off in approx 1.86 hours, whereas the same volume of ethanol would only produce 80 MW and take 14.7 hours to burn-off.

 

Knowing the heat output in MW it is then possible to calculate the radiant heat per square metre at any distance from the source using the inverse-square law. To a first, very rough-and-ready approximation the radiant heat per square metre at some distance from a pool or bund fire can be calculated considering the fire to be a point source located in the centre of the bund at half flame height. This allows for an estimate to be made of the maximum radiant heat per square metre to be made for planning and design purposes. It is important to have some idea of the range of radiant heat values that matter; the following list gives an approximate indication:

•            0.7 kW/m2 – hot summer sunshine

•            1.5-3.0 kW/m2 – cable softening/flaring

•            5 kW/m2 - limit for people to escape; long-term working level in fire-kit

•            12.5 kW/m2 – wood, paper, paint ignite

•            35-40 kW/m2 – steel tank walls deform.

 

Complex corrections have to be applied for flame height, wind direction and emissivity of the smoke cloud in order to achieve more accurate values. Hydrocarbons such as heptanes, xylene or diesel burn with a dark, sooty smoke cloud whereas the smoke produced by oxygenated solvents such as ethanol or acetone is much paler and almost invisible. This affects the amount of radiated heat as black, soot-laden smoke is a much more efficient black-body radiator, with higher emissivity, than pale colourless smoke.

 

In purely practical terms this represents a useful indicator for the officer in charge of the first attendance.

 

A heavy black, roiling smoke cloud indicates a hydrocarbon fuel – quite possibly an aromatic such as gasoline or xylene – requiring AFFF, FP or FFFP foam. On the other hand a pale, almost white smoke cloud with pale blue or near invisible flame would indicate a polar fuel fire and would require an alcohol-resistant type foam, AFFF-AR, FFFP-AR.

 

Quite surprisingly many fire officers when asked how they would determine what type of foam was needed from the character of the smoke, are unaware of this simple rule of thumb. They are also generally unaware of the rule of thumb for identifying whether a synthetic or protein-based foam is in use at an incident; the foam plume for protein-based foams is pinkish or orange-red due to the haemoglobin content of the crude slaughter-house protein (horn-and-hoof) used in its manufacture whereas synthetic foams give a white or cream coloured plume.

 

Strategic planning for the worst-case scenario requires a knowledge of the radiant heat contours for any potential incident. Likely blast over-pressures also need to be calculated, for example using the TNO model[2]. Plume modelling is also necessary to estimate the effects of the smoke cloud. Taken together this information will help to inform policy for evacuation of members of the public and any necessary closure of roads, railways, waterways, or even airspace.

 

Radiant heat and blast information is also essential in determining whether the emergency services can operate safely close enough to the incident – appliances and personnel must be able to reach foam inlet connections and pipework on fixed installations, and foam storage tanks and pumps must not be within the predicted range of damaging levels of blast or radiant heat.

 

Various design faults in storage facilities – tank farms – are identifiable based on historical accident reports or actual pre-incident risk assessment. Currently it is possible to do remote risk assessments using software like GoogleEarth – resolution is perfectly adequate to measure bund areas, tank diameters and spacing, and may be good enough even to see pipework and valve gear. Potential areas of design failure include:

•            Bund volumes which are too small – many are not large enough to hold the entire contents of the ruptured tank(s) within the bund, plus fire fighting foam and any cooling water used operationally;

•            Bund areas are too large, a situation common in large refinery areas, for example, with 80 m x 14 m tanks in a bund 200 m x 200 m – if the area involved in a bund fire exceeds between 6,000 to 10,000 m2 it may be well nigh impossible to extinguish it;

•            Bund wall design must include hydrocarbon or solvent-proof seals between the concrete sections and around pipework passing through the bund wall – the rubber seals between the concrete slabs making up the wall and around pipework failed in the Buncefield incident (see Figure 1);

•            Tank spacing is too small with the risk of neighbouring tank rupture due to radiant heat impingement or blast – this happened at Buncefield. An example of dangerously close tank spacing – 6m or about the length of a living room – is shown in Figure 2 for an airport tank farm;

•            Foam inlets for fixed installations and pumping equipment are too close to be used in the case of a full-area bund fire because of radiant heat;

•            Access roads located too close to bund walls and tanks as regards radiant heat or blast;

•            Tanks sited too close to other hazards raising the risk level – for example, at Amsterdam Schipol the fuel storage tanks are located between an intersection of the main runways on the airfield proper! (see Figure 3.)

 

All of these deficiencies can and should have been avoided at the design stage but if not, then detected and allowed for, at the strategic incident planning stage. Problems may, however, occur as a site expands over the years. Facilities for removing post-incident fire water run-off from bunded areas and their surroundings should also be part of the overall design in order to minimise the potential impact of any incident on the aquatic environment.

 

Sufficient foam supplies held off-site but close by are essential. Powder may also be needed if there is a significant gas risk. Procurement procedures need to identify what is required of the extinguishing agent.

 

For example, does one need good blanketing and vapour suppression qualities? Is extinction time critical? Does the foam have to withstand hot steel surfaces as found in storage tank and process equipment fires? Is seawater compatibility required?

Is an alcohol-resistant (AR) foam for polar solvents necessary? Given the advent of modern biofuels, ie, the high ethanol gasolines or ‘gasohols’ which may contain as much as 85% ethanol, all fire brigades whether municipal or industrial should consider moving over to AR-type foams, especially considering that 100% ethanol is transported and stored for blending.

 

Road incidents involving tankers carrying 100% ethanol will become more common; a recent incident near Huntingdon (UK) on 21st January this year which caused traffic chaos on the A1 and A14 in Cambridgeshire[3] highlighted this issue. Interestingly there was no obvious evidence from the press photographs of a foam blanket having been used to reduce the risk of ignition from any spill.

 

Does one need high-expansion capability (LNG pool fires) or even additionally a Class A foam for carbonaceous fuels, rather than just a Class B foam for flammable liquid fires? Remember that Class B foams are not formulated for Class A use and may fail to penetrate the fuel as happened on 21st August 2009 at a large rubber crumb fire in Littleport East Cambridgeshire which, as a result of no Class A foam being available, burnt for 49 days![4]

 

What approvals or certification does your organisation require for any foam concentrate procured? With storage tank farms the petrochemical industry standard batch test LASTFIRE, together with EN1568, UL162 or NFPA certification are all appropriate. If the site is an airport then ICAO level A/B/C certification will also be necessary, whereas military sites would require Mil-Spec or Defence approvals.

Currently the petrochemical and chemical process industries tend predominantly towards AFFF-AR or FFFP-AR foam concentrates for extinguishing Class B fires. Recent developments in the foam industry include increasing awareness of the potential of fluorine-free foam concentrates, or the use of alternative technologies such as compressed-air-foam systems (CAFS). Many organisations are now feeling pressures to consider the environmental impact of fire fighting operations and are consequently re-evaluating their options for operationally fit-for-purpose Class B extinguishing agents.

 

Finally training must not be forgotten. This too must be fit-for-purpose. Unfortunately many organisations have taken the easy way out under pressure from environmental regulators not to use either real fuel fires or even operational foam for training purposes, which is only subject to a statutory defence against causing pollution in emergencies where human life or health is at risk.

 

Apart from training, this raises important questions tactically for the fire service with incidents not involving ‘persons reported’, such as structural building fires in which only property is at risk. Failure to train properly with operational foam and real fires can lead to potentially serious accidents as a result of those on the front line not understanding how the foam behaves when used, for example, with a large pool fire, especially if they do not have experience of this type of incident. For example, breaking a foam blanket on hot extinguished fuel in a bund by walking through it can result in re-ignition, posing considerable dangers to those involved.

Fire brigades have a legal responsibility under health and safety legislation to provide training that is fit-for-purpose. This means training personnel for situations in which they are likely to have to operate. Common sense would dictate a two-phase approach:

 

(a) Phase 1 could involve training foam used on an LPG-fired rig controlled by the instructor – this provides personnel with experience of handling the foam equipment (yes, it is possible to connect a foam inductor back-to-front or with a smaller hose on the outlet than on the inlet with the result that the Venturi does not work properly, blowing bubbles into the foam concentrate rather than inducting foam! This really happened, thankfully only during a demonstration!);

(b) Phase II training would then be with real fuel, such as gasoline, diesel, biofuels or even pure ethanol, together with operational foam using some form of process rig in a bund, or a contained running pool fire – this second level of training provides direct experience of how real foam and real fuels behave, especially in the presence of hot steelwork forming obstructions around which foam may or may not flow easily, as well as of the problems associated with fuel pick-up using forceful direct foam application as opposed to indirect application off a surface; or of the difficulties associated with highly polar solvents like pure ethanol – ordinary AFFF will not work properly with gasohols containing greater than 10% ethanol. Some Class B fuels like gas condensates in the oil industry are especially difficult to extinguish, requiring considerable experience and expertise, since AFFF-type foams cease to film-form with these low molecular weight volatile hydrocarbons because of their very low surface tension. Direct experience of the high and potentially dangerous levels of radiant heat from a large bund or pool fire are essential – at a foam school recently there were military fire fighting instructors who had never experienced this!

The potential impact on the environment of foam training must be balanced against training outcomes – essentially a cost-benefit-analysis. Whether Phase 1 or Phase II are being considered, foam contaminated fire water run-off must be contained, stored and then treated before discharge according to local regulations. But the bottom line is that foam training must be fit-for-purpose and firmly anchored in real fire scenarios.

 

Notes

1 Babrauskas, V., “Burning Rates,” Section 3, Chapter 3-1, SFPE Handbook of Fire Protection Engineering, 2nd Edition, P.J. DiNenno, Editor-in-Chief, National Fire Protection Association, Quincy, Massachusetts, 1995.

2 B.J. Wiekema, “Vapor cloud explosion model” Journal of Hazardous Materials (1980) 3, 221-232.

3 http://www.peterboroughtoday.co.uk/news/traffic-and-transport/travel_severe_delays_on_a1_after_a14_accident_1_2330649

4 http://www.cambsfire.gov.uk/incidents/3617.php

  • Operation Florian

Sign up: eMagazine & eNewsletter

The latest issues in your inbox.

Company Profiles

HazSim - Bringing situational HazMat training to life

HazSim, LLC provides innovative simulation training to ensure your team works safely and effectively. HazSim Pro simulation equipment is in use by hundreds of fire departments, training schools, industrial fire teams, and private trainers across the US, Canada and further afield as well as the US Army.

Revolutionizing fire fighting foam technology

The one-stop resource for fire fighting foam concentrates and custom-designed foam suppression systems hardware.

Advancing rescue technology

The specialist supplier of quality PPE and Rescue Equipment to Emergency Services.

The ultimate in innovation, quality and service

For 60 years Lehavot has been delivering the world’s most advanced fire detection and suppression automatic systems

Trust the best, let us be your foam solution

AUXQUIMIA is a Spanish company whose main activity is the design, manufacture and commercialization of firefighting foam concentrates.

Williams Fire & Hazard Control offers a full line of specialized fire response equipment for oil and gas platforms

From storage tanks and pipeline emergencies to offshore platforms and vessels at sea, Williams' response personnel and specialized equipment quickly address adverse fire emergencies.

Foam fights fire

Europe’s foremost fire fighting foam manufacturer has been developing and producing foams since the 1920s.

The leader in truck-mounted hydraulic platforms

Our mission is to provide the best and the safest solution to professionals that work at height.

The independent alternative

Dafo Fomtec AB is a privately owned company with head office in Stockholm Sweden and manufacturing in Helsingborg in the south of Sweden.

If you want quality, you want Zico

Since its inception Ziamatic Corp has provided the men and women of the fire service with products designed to make their jobs safer and easier.

Calendar