Quick Links
Customer Centres
Are Mathematical Models an Appropriate Surrogate for Exposure Monitoring when Establishing Respiratory Protective Requirements for the Clean-up of Small Indoor Chemical Spills?
RS2003/04-DG03
Final Report Date: June 2006
| Principal Investigator: | Quinn Danyluk (Fraser Health Authority) and Chun-Yip Hon (Vancouver Coastal Health) | |
| Co-investigators: |
|
For more information about this project, please contact Quinn Danyluk or Chun-Yip Hon.
Issue
When a chemical spill occurs, airborne concentrations and exposure levels are often unknown. As a result, the only option may be to use the highest level of respiratory protection, such as a self-contained breathing apparatus (SCBA). In this project, researchers attempted to develop a way to predict exposure levels resulting from chemical spills using mathematical models, so that exposure monitoring and the use of cumbersome protective equipment could be avoided where possible.
Key findings
- None of the mathematical exposure models tested accurately predicted the actual airborne concentrations that resulted from the spills.
- The mathematical models tested cannot reliably be used to determine the appropriate level of respiratory protection following chemical spills.
Objectives
- To assess the accuracy of mathematical models in predicting the exposure levels that result from chemical spills in the workplace.
- To assess whether these models can be used to determine what level of respiratory protection is required after a chemical spill.
Method
Simulated spills were used to measure actual airborne concentrations for six chemicals commonly found in hospital labs. Various sizes of spills were studied for each chemical, with several repeated trials conducted for each size of spills. Data was collected to calculate the rate of generation (rate at which vapours are produced from the liquid’s surface) and the dispersion rate.
Three models (Nielsen; Hummel; Lennert) were used to predict generation rates and two models (Nicas well-mixed room model; Keil and Nicas dispersion model) were used to predict dispersion rates.
The measured rates of generation and dispersion were compared with the predicted rates to assess the accuracy of the models.
Results
On average, the predicted generation rates were seven times lower than the measured generation rates. The predicted airborne concentration levels based on the dispersion models were also inaccurate in all instances.
None of the mathematical models tested were accurate enough to be used to develop a tool for identifying the appropriate level of respiratory protection for chemical spills in the workplace.
Conclusions
The mathematical models tested are not accurate enough to predict exposure levels for chemical spills in the workplace. One of the limitations of the study was that some of the conditions in the simulated spills, such as air velocity, were different from the conditions in which the mathematical models were developed. However, if the models lose their validity when some basic factors vary, their suitability as a tool for predicting exposure levels in actual spills is questionable.
Future directions
Although the specific models examined in this study were not able to predict actual exposure levels, other models exist and exposure modeling remains a potentially important tool and worthy of future study. The data collected in this study could also be used in future research and could be used to determine the level of respiratory protection needed for spills involving the chemicals and amounts included in the study.
Publications and conference presentations
American Industrial Hygiene Association BC Yukon Section. January 18, 2006. Burnaby, BC.
12th Conference Occupational Hazards to Health Care Workers. September 13-14, 2006. Seattle, WA.
BC Environmental and Occupational Health Research Network’s Annual General Meeting. November 24, 2006. Vancouver, BC
UBC School of Occupational and Environmental Hygiene Seminar Series. December 1, 2006.