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

Abstract View


Spectral Measurements of Mass Absorption Cross-Section of Flare-Generated Black Carbon

BRADLEY CONRAD, Melina Jefferson, Brian Crosland, Matthew Johnson, Carleton University

     Abstract Number: 1481
     Working Group: Carbonaceous Aerosol

Abstract
Gas flaring is a ubiquitous process in the upstream oil and gas industry where unwanted gaseous product is destroyed through open-to-atmosphere combustion. Flaring is a significant global source of atmospheric black carbon (BC), a pollutant second only to carbon dioxide in terms of radiative climate forcing (Jacobson 2001). With respect to the climate, the crucial property relating the mass of BC to its radiative forcing potential is the mass-normalized absorption cross-section (MAC, in units of area per mass). Additionally, MAC is also a necessary variable in optical diagnostics that quantify BC mass from observed attenuation of light. In their well-cited review, Bond & Bergstrom (2006) suggest that a fixed value of MAC should exist for all combustion-generated BC, especially with respect to notable uncertainties in its measurement. The authors also note however, that absorption by BC is sensitive to the degree of graphitization of BC particles, which occurs at elevated temperatures as particles traverse the length of the flame. This inherently suggests a dependence of MAC on the time-temperature history of flame-generated BC. Since buoyancy-driven gas flares are generally much larger in scale than other important combustion-generated BC sources, such as diesel and spark-ignition engines, this notion suggests that the MAC of flare-generated BC (and other sources for that matter) could indeed be different than the value presented by Bond & Bergstrom (2006).

This work presents spectral measurements of MAC of fresh flare-generated BC on laboratory-scale flares representative of the upstream oil and gas industry. Experiments of flares of up to approximately 8' in length are performed at the Carleton University Flare Facility in Ottawa, Canada via vertically-oriented flare stacks spanning 1.5" to 3" in diameter. Combusted flare gas compositions of up to nine components representative of measured flare gases in Alberta, North Dakota, Ecuador, the North Sea, and Russia are combusted over a range of volumetric flow rates. Absorption data from a three-wavelength photoacoustic soot spectrometer (Droplet Measurement Technologies) spanning the visible spectrum is coupled with carbon mass data from a semi-continuous, thermal/optical organic carbon-elemental carbon analyzer (Sunset Laboratory) to directly quantify flare-generated BC MAC. In this study, the focus is to assess the sensitivity of BC MAC to various flare conditions. To this end, fuel flow rates are selected to enable the isolated assessment of BC MAC as a function of flare aerodynamic parameters (e.g. Reynolds number, Froude number, and flame length) in addition to chemical properties defined by flare gas composition.