Abstract View
Repeatable Emission Studies of Controlled Biomass Pyrolysis and Combustion Using a Cone Calorimeter Set-Up
VILHELM B. MALMBORG, François-Guillaume Ide, Dan Madsen, Ioannis Sadiktsis, Michaël Toublanc, Patrick van Hees, Andrew Grieshop, Joakim Pagels, Lund University, Sweden
Abstract Number: 373
Working Group: Combustion
Abstract
Recent work highlights the influence of combustion phase on properties of particles emitted during diverse biomass combustion types. However, systematic study is limited by the highly variable combustion conditions within and across burns. Flow reactor experiments with simple fuels (e.g., benzene) can reduce variability to probe different conditions, but this approach does not directly represent biomass combustion emissions.
In this study, we bridged this gap by using birch wood samples in a cone calorimeter according to ISO 5660-5 to control total heat flux (HF) and gas (air/N2) flow rates to simulate distinct conditions (pyrolysis, well- and under-ventilated combustion). We conducted over 40 experiments with widely ranging HF and flow conditions while monitoring fuel mass loss to quantify emission yields. An Aerosol Mass Spectrometer (AMS), a multi-wavelength aethalometer and a particle size spectrometer (DMS5000) measured time-resolved evolution in particle properties during burns. We also collected filters for offline analysis of OC/EC, PAHs/oxy-PAHs and UV-Vis absorption by methanol extracts.
Pyrolysis conditions (in N2) generated high primary OA emission factors (0.03-0.5 g/g). The Absorption Ångström Exponent (AAE) was 2-3 for pyrolysis conditions, with marginally increased AAEs with higher HF values. For high-flow combustion, we observe increasing eBC emissions (0.005 to 0.01 g/g) and decreasing AAE to values in the range 1.7-1.2 for higher HF, and OA emissions reduced by two orders of magnitude compared to pyrolysis conditions. Low-flow and oxygen-starved combustion was associated with extreme PAH emission factors, up to 0.1 g/g (~100 times higher than under well-ventilated conditions).
Further analysis of the data set will parameterize emissions based on conditions underlying diverse biomass combustion modes and aim to unify observations collected at different scales (e.g., cook stoves, wild fires, heating stoves).