Hazardous area classification and Laboratory operations
This guidance is issued by the Health and Safety Executive. Following the guidance is not compulsory, and you are free to take other action. But if you do follow the guidance you will normally be doing enough to comply with the law. Health and safety inspectors seek to secure compliance with the law, and may refer to this guidance as illustrating good practice.
Hazardous area classification is a technique for assessing the probability of formation of a flammable atmosphere and its likely duration. It has long been a widely used technique in the chemical industry, as a step towards deciding whether electrical and other equipment needs special protective features in order to prevent it causing a fire or explosion. The Dangerous Substances and Explosive Atmospheres Regulations 2002 require a risk assessment of the fire and explosion hazards at a much wider range of workplaces, including laboratories of all kinds. Specifically they require employers to classify areas where explosive atmospheres may form, using the principles of hazardous area classification. No lower threshold for quantities of materials or risk for this is given. Instead, the requirement to designate a hazardous area relates to places where special precautions are needed to protect the health and safety of employees. `Special precautions' are taken to be design features of electrical and other equipment that prevent it creating an ignition source.
In many cases there are health risks associated with exposure to gases and vapours, and laboratory work will be designed to minimise releases, and control concentrations in the air people breath. The concentrations needed to control the health risk are far below the levels at which an explosive atmosphere can form, so precautions designed to protect the health of laboratory workers may also help prevent the need to designate any zoned areas. Nevertheless very localised areas with an explosive atmosphere will often exist, and to prevent an ignition followed by a fire or explosion, the risks must be understood, and suitable controls adopted.
The Regulations define 3 zones that can exist within a hazardous area:
- Zone 0 - A place in which an explosive atmosphere is present continuously or for long periods.
- Zone 1 - A place in which an explosive atmosphere is likely to occur in normal operation occasionally
- Zone 2 - A place in which an explosive atmosphere is not likely to occur in normal operation, but if it does occur, will persist for a short period only
For most laboratory operations there is no tradition of hazardous area classification or using ignition-protected equipment, and the risks are usually controlled in other ways. Nevertheless, fires and occasional explosions do occur, and many of the principles used on larger scale operations to control the risks are relevant, even if the solutions are different. The purpose of this document is to help those in control of laboratories to carry out the risk assessment required by DSEAR, and to identify the circumstances in which controls to prevent the formation of explosive atmospheres should be improved, or the area should be classified.
Where an area is classified, DSEAR requires electrical and non-electrical equipment to meet specific standards. It is helpful to analyse the risks systematically when reviewing if existing equipment and systems of work are adequate.
Area classification analyses the sources of gas and vapour release, looking at those that arise in normal operation, (primary sources) and those which only occur as a result of some foreseeable equipment failure or operator error (secondary sources). Clearly, not all laboratory hazards will be covered by such an analysis, but it should be part of any overall risk assessment. The aim should always be to minimise releases into the general atmosphere of the laboratory.
On a laboratory scale, a primary source might be the vapour released when a volatile solvent is poured from one container to another, while someone knocking the container off the bench to the floor creates a secondary source. Primary sources should normally be so small they can easily be controlled by adequate ventilation, and the extent of any explosive atmosphere is negligible.
Strictly, area classification takes no account of the consequences of a release, whether this is a fire or explosion, but selection of the necessary precautions must take the consequence factor into account and this approach is recognised in DSEAR. If precautions already used are adequate to prevent fire and explosion risks to laboratory workers, then there is no need for zoning and `special precautions' in terms of the ignition risk from equipment will not be necessary.
For laboratory work, it is helpful to consider separately releases which occur suddenly, but where the maximum release quantity can be specified, e.g. the fracture of a glass flask; and those where the release once started will continue until some corrective action is taken to shut off the release, e.g. closing a valve following failure of a plastic or rubber hose from a gas cylinder or mains supply.
The volatility of the product is also an important factor, but this must be considered in the context of the temperature at which it will be used. So if you are pouring a solvent from one open container to another at a temperature below its flash point, there should be no hazardous area, because too little vapour is present. If you then distil the same solvent and the condenser cooling fails, vapour will be released, and the risk is much greater. Consequently, the comments on flammable liquids apply whenever they are used or stored above their flash point.
The degree of supervision of any continuous process should be considered. Would someone be constantly available to take action, if cooling water flow failed, a flask cracked, or a process boiled over? Would it be possible to isolate all electrical equipment quickly from a safe place? Turning off switches that are in the immediate vicinity of a release might create the spark that we need to avoid. Familiarity with an operation carried out many times should not be the basis for leaving the equipment unsupervised, if some simple problem could create a risk to others.
Training is also relevant to assessing the risk. Students and other young people may not have the experience in laboratory work that fully experienced staff have, but they can and should be given training in the actions that should be taken to prevent foreseeable problems, like small spills, and the action needed where simple problems arise.
Most laboratories have good general ventilation, but this is not primarily intended to limit the extent of any explosive atmospheres that may form. More localised extraction is needed for this.
Very small scale operations
Looking first at operations with flammable liquids, at the very smallest scale of operations, the consequences of a spill may well be trivial. Quantities up to about 50mls can be mopped up or sometimes flushed away, and if they ignite, so long as the fire does not quickly spread, they may well burn out before anyone is at risk, or before a laboratory worker could take any action to extinguish a fire. If these are the conclusions of a risk assessment, formal zoning is clearly inappropriate, though it may well be appropriate to avoid the use of naked flames and other powerful or constant ignition sources in the immediate vicinity. Where the evaporation of a solvent is deliberately intended, e.g. from a coated surface, the operation may need to be carried out in a fume cupboard. In these cases, if the health risks under COSHH are properly controlled, there may well be no need for additional precautions to control the fire and explosion risk.
It is expected that most work by students or pupils in schools will be at this scale.
Medium scale operations
Where quantities are larger but still manipulated on the open bench, for example up to 2.5 litres, in a Winchester bottle, the risks are more significant. The actual extent of a flammable atmosphere following a spill may well be a radius of up to a metre, but only a very small height above the liquid level. Any ignition of a spreading pool will produce a fire that quickly extends to the whole area of the spill, and could cause a risk to laboratory staff. Particular dangers arise if the spill enters the drains, as an explosive atmosphere could then form in an enclosed space.
Direct heating by bunsen burners and other obvious continuous ignition sources should be avoided, but the greatest risk probably comes from electrical equipment in use as part of the operation. Much of this cannot be avoided, and may well not be available in ignition-protected form, e.g. hot plates, heating mantles, stirrer controllers.
Precautions are likely to include: good handling techniques to minimise spills, sills or other liquid retaining methods to minimise liquid spread, proper support for glass equipment, placing electrical equipment where it will not be splashed as a result of a spill as far as practicable, and constant supervision by trained staff, so that electrical equipment can be rapidly isolated, others warned of any dangers, and if safe to do so, first aid fire fighting started. Most importantly, the risk that a fire following a spill will rapidly involve other containers of flammable liquids or other dangerous chemicals should be considered. Any necessary improvements to the storage arrangements for such chemicals should then be implemented. Where these and similar precautions have been adopted, the risk assessment may conclude that there is no need for hazardous areas to be specified.
Instruments with internal flames
Some types of instrumentation use very small internal gas flames, and in principle could ignite any surrounding explosive atmosphere, if for instance there was a release of vapour from some operation nearby. In addition, an explosive atmosphere could form from a leak in the fuel line to the flame, or in some cases from flammable liquids in the instrument. Such instrumentation needs specific consideration in the risk assessment. If the maximum size of leak is very small, any release will form an explosive atmosphere of negligible extent.
Some work at this scale may be done in a fume cupboard, and this will allow the sash to be closed to give some protection if a fire should start. The work should be arranged so that any foreseeable release of gas or vapour will be rapidly diluted below the explosive limit, by the airflow through the cupboard. Precautions may still be needed to reduce the fire risk, such as retaining sills at the front edge, and extraction ductwork kept free from flammable residues. In particular fume cupboards should not be used as storage facilities for toxic or flammable chemicals while they are also being used for experimental work. Rapid failure of stored bottles in a small fire could produce sufficient vapour to prevent the extract fan diluting vapours sufficiently.
Liquefied flammable gases
These may be handled either under pressure, or in refrigerated form. A small release of liquid is likely to vaporise immediately, creating a substantial size of explosive gas/ air mixture. Pressurised systems need to be robustly constructed, and checks provided to ensure they are leak tight. Where liquids are handled in refrigerated form, the risks from loss of cooling or loss of insulation should be considered. Good ventilation around the apparatus will always be needed, but there may also be a need to designate a zone 2 area. This will depend on the foreseeability of a release of liquid, how rapidly it might be detected, and the ability of the ventilation to disperse it quickly.
Leaks that continue until some corrective action is taken are possible from some piped gas sources. Low pressure mains gas pipes in domestic and similar premises are not considered to give rise to hazardous areas although explosions caused by gas escapes in private houses do occur from time to time. Laboratory work presents a wider range of hazards:
- Many gases used are not odorised; some may be lighter than air, while others are heavier,
- Gas pressures often are significantly higher than the 75mbar pressure typical of mains gas,
- Many temporary connections are made using flexible rubber or plastic hose which can fail with age, or from unsuitable connection methods or attack from propane, butane or solvents. Equipment could fail from overpressure, if a regulator is wrongly adjusted. Some types of gas taps on open benches can easily be knocked open inadvertently
The risk assessment needs to consider: the ventilation provided, its reliability and the size of leak that could be controlled in this way; what can be done to minimise the risk of a gas leak, and how any such leak would be identified promptly, so that appropriate action could be taken. Particular account should be taken of the out-of-hours risk if ventilation is shut down, or systems have to be left under pressure.
Small leaks may well be dispersed safely by good ventilation, but the consequences of a release that builds up and then finds a source of ignition are likely to be severe.
Releases into enclosed spaces
Where vapours or gases may escape into an enclosed space like an oven, or refrigerator, the consequence of an ignition is more likely to be an explosion than a fire. Refrigerators have exploded in laboratories, where the light switch or thermostat contact sparked when opening or closing. This risk can be avoided by buying a unit designed for this purpose, with any spark-producing electrical equipment sealed from contact with the internal atmosphere, rather than a designation of the inside of unit as zone 1 or 2. In the case of an oven, it may be possible to keep heating elements below the ignition temperature of any vapour likely to be used, or to provide adequate ventilation to prevent the build up of vapours, but some risks will remain unless close control is maintained over products and quantities that can be placed inside.
Larger scale laboratory work
Laboratory work involving equipment above a 2 litre scale, and pilot scale plants need more careful consideration. Pilot scale is taken to mean equipment with a capacity of 50-100 litres or more.
Particular risks come from the use of all glass equipment that may be fractured by impact, thermal shock, overpressure, assembly technique, and poor handling with open containers, use of temporary hoses for flammable or other hazardous materials.
Perhaps the most useful approach to controlling the ignition risks, is to limit the extent of any flammable atmosphere formed as a result of a release, by a combination of semi-enclosure, forced ventilation, and then to place all electrical equipment outside the enclosure, so far as possible. This may allow a hazardous area study to conclude that any zones are of very small or even negligible extent. Direct heat sources, like an electric mantle may nevertheless need to be used, and could be exposed to flammable vapours following a major failure of a glass vessel. In this case the risk assessment should consider if the laboratory worker and any others nearby could be expected to escape safely, and how any subsequent fire could be prevented from spreading to affect other people and facilities.
Where flammable liquids or gases are handled in laboratories, it is always necessary to consider sources of ignition in any risk assessment, and control these closely, even if there are no formally designated hazardous areas.
Whether work is done on the open bench, in a fume cupboard or a dedicated facility for larger scale work, a decision in a written risk assessment not to zone the laboratory must justify this on the basis that any incident will be of limited scale and could be safely and quickly controlled by those present; or that they could escape very quickly without leaving others in the building at risk. A high standard of controls of the type described above will help justify this assessment.
There are circumstances, in particular at the pilot scale, where hazardous area classification, and the associated use of explosion protected (Ex) equipment should be adopted. In all cases, the design of the facility should aim to minimise the extent of any hazardous area if it cannot be avoided completely.
Other sources of information
Royal Society of Chemistry, note from the environment health and safety committee on Fire Safety in Laboratories.