skip to primary navigationskip to content

Asphyxiating atmospheres

Incidents With a Risk of Asphyxiation

This page covers the following topics:

The risks of asphyxiation in the Department

The principal risks of asphyxiation in the Department of Physics are:

  • Breaches of the insulated liquid nitrogen delivery lines (in the green stairwell) **
  • Leaks from the nitrogen gas lines - these are fed from the liquid nitrogen tanks and, when full, they contain sufficient to deliver of order 25 000 cubic metres of gas (at RTP) to the Mott building, and 6 300 cubic metres to Microelectronics/Kapitza. In Physics of Medicine they are fed from a supply generated from the air, so this is unlimited. **
  • Dewar leaks/failures **
  • Quench from a magnet. This can vary from a negligible risk to 800 cubic metres of gas per minute for a large NMR magnet.

** Note: in the last eight years we have had incidents of each of these, and any of them could potentially have led to loss of life. Please take this topic very seriously.

Misconceptions about oxygen deficient atmospheres

If you read the literature you will find tables listing the signs and symptoms of oxygen deficiency. However, it cannot be stressed too much that a person in an oxygen deficient atmosphere will be unable to detect these signs and symptoms in themselves. Indeed they will feel completely in control of the situation, even up to the moment before they die from the effects.

This was demonstrated graphically by a film made for the Horizon series by the BBC of a well known politician being taken within seconds of asphyxiation. He was unable to subtract one small number from another and unable to complete a sorting exercise designed for a two-year old. He also repeatedly disobeyed the direct orders to 'put up the switches or you will die'. He was rescued by the direct intervention of Hans Wittenberg, the person accompanying him. Afterwards he was confident that he had done extremely well. He stated 'I fooled you, didn't I?', to which the response was 'No, Michael, death was your final destination today, Hans saved your life". You can see a brief extract from this film 'How to kill a human being' (the relevant part of this clip starts at 6.30). Clearly oxygen-deficiency has profound effects on the brain and our perception of our capability.

The second misconception is that you can hold your breath and go in to fix the problem or rescue a co-worker. You can't. Industrial accident records show that on average an asphyxiating atmosphere will claim two or more lives. One or more will be would-be rescuers.

The third misconception is that people who are rescued will recover. This is unlikely; once someone has become unconscious from lack of oxygen they are unlikely to recover even if they are rescued.

The fourth misconception is that the percentages in the tables are correct. In particular you should question the accuracy of the threshold between life and death. For obvious reasons the data cannot be reliable.

How to reduce the risks

Liquid nitrogen delivery lines

These are fixed and not normally expected to fail. Failures in the vacuum jacket should be taken as a sign that there is a serious problem which should be fully investigated. It is possible that the nitrogen is leaking into the jacket - left to its own devices this may eventually cause a catastrophic failure and liquid nitrogen fountain. Ensure that someone knows how to turn the supply off from the source.

Nitrogen gas lines

There is a network of pipes within the lab. Plans of these pipes are kept in the emergency files. To reduce the risk of asphyxiation from these lines it is imperative that good quality joints are used at all times.

Brazed or welded joints are good. Swagelock or similar fittings are also good.

However push-on tubing with, or without, jubilee clips or pieces of twisted wire, is unacceptable.

brazedjoint swagelock unsafejoint

People who work in rooms with a nitrogen supply need to be aware of how to shut off the supply in an emergency - this should include a knowledge of a way to shut it of from OUTSIDE the room, since it may not be safe to enter in an emergency.

If it is reasonably foreseeable that nitrogen might be released in the room then oxygen monitoring should be considered. This may either be a fixed mains-operated monitor or a portable monitor. Monitoring is only successful if the equipment is kept in good working order. Please ensure that it is maintained - otherwise it may not save someone's life. You may wish to use the oxygen calculator to help you to decide whether oxygen monitoring is needed.

If there is oxygen monitoring in the room, it is good practice to make it audible or visible outside the room.

If there is oxygen monitoring in the room, there should be the standard emergency notice posted prominently outside.

Dewar leaks/failures

All users of dewars should be taught the layout of the valves and how they should be left. This will reduce the risk of a dewar misbehaving. All users of liquid helium should be taught how to leave the dewar in a safe condition.

Dewars should be inspected and maintained according to the schedule, and this is overseen by the Liquid Gases technician. Dewars whose (volume x the rating of the pressure release valve) exceeds 250 bar litres must be on the insurance schedule and inspected by the insurance engineer. Every dewar should carry a label that indicates its status as regards inspection and maintenance.

Magnet quench

It is essential to consult the manufacturer's literature, or other information, to ascertain the maximum rate at which gas could be released.

Armed with that data you can calculate whether the release will threaten the well-being of people working in the room. You may wish to use the oxygen calculator to help you to do this.

If the result of the calculation leads you to believe that it could cause an asphyxiating atmosphere the best course of action is to install a quench pipe, to pipe the released gas straight out of the building. If you decide to install emergency ventilation you need to address the following aspects:

  • what is the flow rate of escaping gas?
  • how quickly does the room become dangerous?
  • how quickly does the emergency ventilation system react?
  • how well can it cope with the flow rate of escaping gas?
  • should it be extracting from high level or low level?

Failure to address all of these points could result in an emergency ventilation system that does not start up quickly enough, fails to extract gas rapidly enough or pulls the good air out of the room before the cryogen.

Note that oxygen monitoring may be required, but more to tell you when the room is safe to re-enter.

How to handle an incident

How to use an oxygen meter

First check that it is working - it should read approximately 20%. If it does not, then re-calibrate it. Check that when you breathe over the oxygen sensor the reading goes down and the alarm sounds. If it does not then the equipment is not working correctly and you need to find another meter.

The meter reacts relatively slowly to changing concentrations, so do not move too quickly. A slow, steady walk is best. Hold the meter in front of you, do not have it hidden under your clothing. Do not breathe the gas coming through the doorway; if it is seriously oxygen deficient it could prove fatal.

If you wish to enter a room, open the door very slightly and put your arm and the meter round the door. If the meter alarms, do not enter.

Do not make judgements on whether a room is safe to enter dependent on the reading on the meter and a table of symptoms you have read somewhere. In particular, be aware that most meters do not give true readings when the asphyxiant gas is helium - they are only linear for nitrogen. Most meters will OVER ESTIMATE the oxygen level, when the cryogen is helium.


What to do if a room oxygen monitor is sounding or there are suspicious 'hissing' noises

asphyxiation decision tree

Click on the image to the left to get a Powerpoint file for this notice. Please ensure that, if you have a monitor in your room, this notice is posted on the door.

If you see a casualty collapsed on the floor, under NO circumstances should you attempt a rescue. The Fire Service have the equipment to do this.

If there appears to be a gas or liquid release, and you cannot figure out how to stop it, then call Jane Blunt.


Dewar leaks/failures

If a dewar falls over, it should be inspected for damage. Rough handling tends to lead to failures at the neck of the inner vessel, which will eventually lead to the jacket plug being ejected. If this happens, the dewar should be approached with caution (holding an oxygen meter) and, if safe to do so, the dewar should be taken outside. If this is not possible, arrangements should be made to ensure that no-one enters an oxygen deficient atmosphere caused by the gas venting. If this happens indoors, then it will be necessary to check nearby rooms, and test to see when the gas has dispersed; this could take some hours.

If a dewar starts to vent its contents, approach with caution, preferably using an oxygen meter to check for danger. If it is not possible to rectify the situation, ensure that the gas is dispersed and that no-one enters the danger area.

Magnet quench

The measures to take should already have been spelled out in your working procedures.

This page was lasted updated 22 October 2013