10th International Aerosol Conference
September 2 - September 7, 2018
America's Center Convention Complex
St. Louis, Missouri, USA

Abstract View


Interpreting Emissions from Biomass Cookstoves and Wood Stoves Using a Simple Pyrolysis Model

Christina Andersen, Robert Lindgren, Ricardo Carvalho, Vilhelm B. Malmborg, Erik Ahlberg, Natxo Garcia-López, John Falk, Axel C. Eriksson, Thomas Kristensen, Birgitta Svenningson, Christoffer Boman, JOAKIM PAGELS, Lund University, Sweden

     Abstract Number: 1478
     Working Group: Combustion

Abstract
Air pollution poses a challenge to both health and climate. In the developing world, 3 billion people use solid fuels for cooking in traditional open fires and stoves, which are major contributors to household and ambient air pollution (Landrigan et al., 2017). However, detailed particle emission characteristics are missing for these types of stoves.

As part of The Salutary Umeå STudy of Aerosols IN biomass cookstove Emissions (SUSTAINE), cookstoves of different technological advancement; a 3-stone fire, a rocket stove (both using wood sticks), and a natural draft (ND) gasifier stove (using wood pellets), were tested using an adjusted procedure of the standardised water boiling test. The wood sticks were from agroforestry cultivated energy wood in Kenya, while the pellet fuel was based on typical Nordic softwood (pine and spruce).

We performed highly time-resolved analysis of transient emissions. Additionally, full batch emissions were transferred to a 15 m3 stainless steel chamber to investigate the impacts of atmospheric aging by passing the aerosol through an oxidation flow reactor (PAM [OH] ~ 1 × 108 molecules cm-3 h-1). A SP-AMS was used to measure the non-refractory and refractory particle composition including particle phase PAHs. A seven wavelength aethalometer was used to derive equivalent black carbon (eBC) mass concentrations. Major combustion gases were measured with on-line FTIR spectroscopy. Emission factors (EF; in units mg/MJfuel) were calculated on the basis of AMS, aethalometer and CO2 measurements. The emissions were compared to emissions characterised from a conventional Nordic wood stove for domestic heating (Nielsen et al., 2017).

The emissions were interpreted in a framework of basic knowledge from wood pyrolysis (Milne 1989), where the temperature in the fuel gases is a controlling variable. The results show that the least advanced cookstoves (the 3-stone fire and the rocket stove) are associated with the highest emissions of primary organic aerosol (POA), dominated by anhydrous sugars and methoxy phenols in the particle phase and moderate emissions of benzene, toluene and xylene (BTX) in the gas-phase relative to POA. We interpret this as a result of lower effective pyrolysis and fuel gas temperature, compared to the combustion conditions in the more advanced stove. More specifically, the PM emission factor was reduced and the rBC/OA ratio was higher for the Rocket stove compared to the 3-stone fire. Furthermore, the ND pellet-fired cookstove showed even more reduced POA and PM1 mass emissions but high emissions of BTX relative to POA and genotoxic compounds such as 5-7-ring PAHs. This led to higher OA enhancements upon processing compared to the less advanced stoves. These observations are consistent with what can be expected by an increased effective pyrolysis temperature.

The conventional Nordic wood stove operated at nominal burn (NB) rate had the overall lowest emissions of POA, BTX, PAHs and SOA. This is also consistent with what can be expected by a higher combustion temperature in the heating stove, compared to the cookstoves. When operated at high burn rates, strongly increased BTX and PAH emissions were found (Nielsen et al. 2017), which explains very high OA enhancements (factor ~10 increase in OA upon aging), presumably explained by a reduced pyrolysis temperature compared to NB. The rBC emissions were still rather high for the wood heating stove, which we hypothesize was due to inefficient soot oxidation (removal) processes.

In conclusion, clear differences in PM1 emission factors of rBC, POA and SOA precursors were seen from different biomass cookstoves and one conventional wood stove for domestic heating. We suggest this could be interpreted by previously defined fundamental knowledge of wood pyrolysis.