The effects of modified operation on emissions from a pellet-fed, forced-draft gasifier stove

Abstract

Traditional solid fuel cookstoves emit gas- and particle-phase pollutants that contribute to household air pollution, human disease, and climate impacts. Forced-draft semi-gasifier stoves are an attractive intermediate step to zero-emitting stoves due to their reported lower emissions in laboratory and field studies, and potential for increased availability in more rural locales. However, emissions from these stoves have been shown to be highly variable and sensitive to stove design, fuel type, secondary air velocity, and operation mode. We measured carbon monoxide (CO), particulate matter (PM2.5), organic and elemental carbon, and particle number (15–685 nm) emissions of the widely adopted Mimi Moto pellet-fed, gasifier stove for different operating conditions under two modified protocols, the Water Boiling Test (WBT) and an updated laboratory testing protocol ISO 19867-1 (ISO). We categorized operating conditions into three approaches: Startup (varying ignition material), Shutdown (varying fan speed during a 45-min burnout period), and Refuel (varying the height of charred pellets added for re-ignition). Refueling led to the largest and most variable emissions, but lab emissions were all lower than high field emissions (e.g., similar to those of traditional solid fuels) and remained primarily in ISO Tiers 5 and 4 for CO and PM2.5, aspirational and second-best, respectively. We find large relative differences in emissions when comparing our results to similar studies conducted with the Mimi Moto and ISO protocol, suggesting small operational differences can have large emissions implications. To minimize emissions, we recommend using kerosene for ignition, turning the fan off when pellets are done burning and flame has extinguished, and reigniting with fresh pellets instead of pellet char. Improved training and maintenance are needed in real-world applications to decrease the frequency of high-emission events. Tightly constrained testing and detection limits remain challenges to fully understanding factors contributing to these events.