Comparison of carbon monoxide levels during heating of ice and water to boiling point with a camping stove.

Objectives To determine whether using a camping stove to bring a pan of ice to boiling point produces higher carbon monoxide (CO) concentration than would bringing a pan of water to boiling point. The hypothesis was that ice would cause greater CO concentration because of its greater flame-cooling effect and, consequently, more incomplete combustion. Method This was a randomized, prospective observational study. After an initial pilot study, CO concentration was monitored during 10 trials for each of ice and water. A partially ventilated 200-L cardboard box model was developed and then used inside a chamber at −6°C. Ice temperature and volume, water temperature and volume, pan size, and flame characteristics were all standardized. Temperature of the heated medium was monitored to determine time to boiling point. Carbon monoxide concentration was monitored every 30 seconds for the first 3minutes, then every minute until the end of each 10-minute trial. Results There was no significant difference ( P >.05) in CO production levels between ice and water. Each achieved a similar mean plateau level of ∼400ppm CO concentration with a similar rate of rise. However, significantly higher ( P =.014) CO concentration occurred at 4 and 5minutes when the flame underwent a yellow flare; this occurred only on 3 occasions when ice was the medium. Conclusion There were no significant differences for CO production between bringing a pan of ice or water to boiling point. In a small number of ice trials, the presence of a yellow flame resulted in high CO concentration. Yellow flares might occur more often with ice or snow melting, but this has not been proven.