February 2008

The DOE Office of Emergency Management (OEM) has issued the following Emergency Management Update relating to the withdrawn of Oxygen Temporary Emergency Exposure Levels. Using the Emergency Management Issues Special Interest Group (EMI SIG) Listserv, OEM periodically sends updates related to emergency management topics of interest.

Previous issues of updates are available on the EMI SIG website and on the Subcommittee on Consequence Assessment and Protective Action (SCAPA) page. Generally, in cases where the subject requires more than one page of discussion, the update includes a summary discussion and links are provided for more detail. This current update is an exception to that rule.

Oxygen TEELs Withdrawn

HISTORY

TEELs for “Oxygen (liquid)” (hereafter referred to simply as “Oxygen”) were developed and introduced in the Protective Action Criteria (PAC)/Temporary Emergency Exposure Limit (TEEL) Tables for Revision 18 based on a request from a Department of Energy (DOE) PAC/TEEL user. These values were:

The TEEL-3 was based on animal (mouse) lethal toxicity data for inhalation of pure oxygen over a 24-hour period, and the TEEL-2 (typically used for emergency planning and response) was based on human effects data taken from a paper in the Journal of the American Medical Association describing administration of pure oxygen for 14 hours to a patient that caused respiratory tract irritation. In the absence of specific data for TEEL-1 and TEEL-0, these values were estimated from the TEEL-3 and TEEL-2 data using professional toxicology judgment. These Oxygen TEELs remained in place through PAC/TEEL Revisions 19 and 20.

Although the above TEELs were set in the manner described, the following additional oxygen toxicity considerations make it difficult to set reliable limits based on effects other than asphyxiation due to the lack of oxygen:

• Although the TEEL-3 was based on mouse lethality data over 24 hours and lethality from inhalation of pure oxygen has been observed in other animal species over similar short time periods, only mild irritation is usually seen in humans after about 24-72 hours of continuous inhalation of pure oxygen.
• In humans, it usually takes 3-10 days of pure oxygen administration to begin to see moderate fluid build-up (edema) in the lungs.
• In addition, there is a wide range in time for human adults when mild respiratory irritation occurs.
• Oxygen toxicity is now recognized to be due to a transformation in biological systems of normal oxygen to a superoxide free radical scavenger that, among other effects, may play a significant role in causing lung irritation and edema.
• Superoxide formation is also found to varying degrees in the oxygen supplies themselves, which may help explain the wide variation in human response.
• Oxygen is an unusual chemical since it is needed for life. As a result, asphyxiant effects appear at relatively low levels of oxygen but adverse effects are nonexistent over quite a large intermediate range. However, at very high oxygen concentrations, toxic effects again appear.
• The 14-24 hours or more exposure to pure oxygen to cause mild lung irritation in humans is extremely unlikely in an emergency release event.

DISCUSSION

Starting with Revision 21, a uniform set of TEELs was established for all simple asphyxiants (such as Helium, Argon, Xenon, Hydrogen, etc.). These values were based on the toxicity observed from oxygen displacement and deprivation. This is an effect that all simple asphyxiants display as a group. OSHA’s lower limit for oxygen in a confined space is 19.5% oxygen. This concentration provides an adequate amount of oxygen for most work assignments and includes a margin of safety. If the concentration of a simple asphyxiant reached about 60,000 parts per million (ppm), it would displace enough air to reduce the oxygen concentration to the OSHA lower limit (i.e., 19.5%). This was used to set the TEEL-0 limit. Similarly, the toxic effects of oxygen deprivation due to asphyxiant displacement of oxygen were used to set the other TEEL limits. In Revision 21, the introduced TEEL values for simple asphyxiants were:

At that time, the Simple Asphyxiant TEELs were set as upper limits so that no chemical could have a TEEL higher than these values. The rationale for this was that if a chemical had no significant toxic effects, then it would be limited to its ability to displace oxygen. However in retrospect, these ceiling limits set up a conflict with the Oxygen TEELs which are higher than the TEELs for simple asphyxiants. This resulted in unintended modifications to the Oxygen TEELs. The changed Oxygen TEELs were unnoticed at this time since no deliberate changes had been made to them. Therefore, they were not on the quality assurance review schedule. This conflict, in combination with other factors in the TEEL development methodology software macros, led to the following incorrectly modified Oxygen TEELs in PAC/TEEL Revision 21:

After discussions with the DOE Office of Emergency Management and the American Industrial Hygiene Association Emergency Response Planning Committee who develop Emergency Response Planning Guidelines (ERPGs) about modifying application of the toxicology involved, changes were made to the Simple Asphyxiant TEELs resulting in the TEELs for Revision 22:

Again, the Oxygen TEELs were incorrectly modified by the Simple Asphyxiant TEEL ceilings, yielding the following TEELs in PAC/TEEL Revision 22:

A short time after PAC/TEEL Revision 22 was released, one of the members of the PAC development team noticed the contradiction in the Oxygen TEELs. The TEEL-0 (3.5% oxygen) and TEEL-1 (6.5% oxygen) that are intended to be a no-effect and minimal-effect level respectively were obviously lethal oxygen deprivation concentrations.

After recognizing that the changed Oxygen TEELs represented a contradictory condition, the PAC development team began examining the TEEL development software and at the same time questioned the need for Oxygen TEELs.

CONCLUSIONS

A spill of liquid oxygen would result in an increase in the oxygen concentration in the surrounding air. The increased oxygen concentration that would likely be observed in a release of this kind would typically not have the toxic properties associated with breathing pure oxygen for several hours, the basis for the TEEL-2. In addition, there are greater risks associated with the increase in flammability created by the increase of available oxygen for combustion when an ignition source is introduced.

Therefore, the PAC development team recommended and the DOE Office of Emergency Management approved withdrawing the Oxygen TEELs in PAC/TEEL Revision 23. Emergency management of this kind of a release should focus on the much greater risks involved with the flammability and ignition source issues.

As additional information for planners, the various Simple Asphyxiant TEEL concentrations that displace oxygen are associated with the residual oxygen levels shown below as the flip side of the coin to the Simple Asphyxiant TEELs noted above in PAC/TEEL Revision 22. They are derived from the Simple Asphyxiant TEELs using simple mathematics and are likely more reliable than TEELs based on other toxic effects mentioned in the History section.

The equation used is as follows:

C_o = ( 1- C_x ) 0.209
where
C_o = Content of oxygen remaining in the air, unitless fraction
C_x = Content of introduced contaminant gas, unitless fraction (e.g. 500,000 ppm = 0.5)
0.209 = Fraction of oxygen in normal air

Notes:

1. Values in ppm are divided by 1,000,000 ppm to arrive at an air fraction. Conversely, air fractions are multiplied by 1,000,000 ppm to arrive at ppm.
2. This assumes a one for one mechanical displacement of air molecules by contaminant molecules with no other molecular interactions or chemical reactions occurring.

These Oxygen TEELs have never been published in the PAC/TEEL Tables but are discussed in Section 3.3.1 “Simple Asphyxiants” of the draft Temporary Emergency Exposure Limits: Methods and Practice that is currently on the DOE REVCOM review website and will be published as a DOE Handbook. The handbook discusses in detail how TEELs are developed: