Review of recent literature on the light absorption properties of black carbon: Refractive index, mass absorption cross section, and absorption function

Abstract

Knowledge of the optical properties of soot black carbon (BC) is required for the prediction of the radiative effects of freshly-emitted and aged BC particles. Here we review BC mass absorption cross section (MAC) and absorption function E(m) measurements, focusing on freshly-emitted BC. First, we review recently reported MACs at 550 nm wavelength as obtained from direct measurements of particulate absorption and mass concentration; we find an average of 8.0 ± 0.7 m²/g from ten measurements, not significantly higher (p > 0.26) than the widely used MAC of 7.5 ± 1.2 m²/g recommended by Bond and Bergstrom [Bond, T. C., and R. W. Bergstrom. 2006. Light absorption by carbonaceous particles: An investigative review. Aerosol Sci. Technol. 40(1):27–67]. Second, we review recently reported E(m), whose retrieval is more complex due to the need to combine measurements with numerical models to estimate the contribution of scattering to extinction. Third, we review recent numerical studies that have aimed to predict the BC MAC using various complex refractive indices (m = n + ik). Most of these studies have used m = 1.95 + 0.79i recommended by Bond and Bergstrom (2006), yet failed to predict a MAC as high as 7.5 or 8.0 m²/g at 550 nm wavelength. Fourth, we summarize a selected range of alternative values of m that has been reported by recent studies and place them in the context of measurements using a contour plot of E(m) on the n–k plane. We show that the widely used m = 1.95 + 0.79i corresponds to an E(m) that is too low to be consistent with the measured MAC values. We conclude that the E(m) of BC in the visible and near infrared should be greater than 0.32, and that the commonly used BC models or the refractive index, or both, are still in need of improvement.

EDITOR:

• Matti Maricq

Acknowledgments

Part of this work was performed on computing resources provided by CRIANN (Normandy, France).

Funding

The financial support of this work by NRCan PERD through TR3 Projects 3B03.0002B and EIP Project EU-TR3-04A is greatly acknowledged.