Field studies on the ammonia odor threshold based on ambient air-sampling following accidental releases

Abstract

Ammonia is one of the most widely produced and utilized chemicals in the U.S., and while numerous lab studies have been undertaken, there is no consensus on the odor threshold. In contrast to controlled lab conditions used to determine odor thresholds, the field conditions following unintentional chemical releases are uncontrolled and highly variable due to many factors. A critical component in managing the response to these chemical spills involves understanding how lab data could be applied to uncontrolled field conditions in and around the affected community. It was postulated that analysis of field data collected following accidental releases of ammonia might augment and verify data collected in lab experiments. The widespread transport and use of ammonia has resulted in a number of unintentional releases of ammonia into the environment as a result of train derailments, tanker spills, and plant accidents. In the field studies reported here, air monitoring data were collected following a variety of accidental ammonia releases. Of 6539 readings between 0 and 1 ppm, odor was detected only in 208 samples (3.2%). Of 65 readings between 1.1 and 1.5 ppm, odor was detected in 51 samples (78.5%). These data are consistent with an ammonia odor threshold within a concentration range of 1.1–1.5 ppm. This level is consistent with the recently published odor threshold data for ammonia, but is significantly lower than frequently cited historical data. Furthermore, a review of the ammonia literature demonstrates that the ammonia odor threshold is significantly lower than levels that produce eye, nose, or throat irritation.

Keywords:

• ammonia
• odor
• field studies
• accidental releases

Acknowledgments

The authors have served as litigation consultants on matters related to ammonia toxicity. Some of the discussions in this article were developed during the litigation process. The interpretation of the data presented in this review, as well as the conclusions reached herein, are the sole opinions of the authors.