Comprehensive characterization of humic-like substances in smoke PM<sub>2.5</sub> emitted from the combustion of biomass materials and fossil fuels

Abstract

Abstract. Humic-like substances (HULIS) in smoke fine particulate matter (PM 2.5 ) emitted from the combustion of biomass materials (rice straw, corn straw, and pine branch) and fossil fuels (lignite coal and diesel fuel) were comprehensively studied in this work. The HULIS fractions were first isolated with a one-step solid-phase extraction method, and were then investigated with a series of analytical techniques: elemental analysis, total organic carbon analysis, UV-vis (ultraviolet-visible) spectroscopy, excitationemission matrix (EEM) fluorescence spectroscopy, Fourier transform infrared spectroscopy, and 1 H-nuclear magnetic resonance spectroscopy. The results show that HULIS account for 11.2-23.4 and 5.3 % of PM 2.5 emitted from biomass burning (BB) and coal combustion, respectively. In addition, contributions of HULIS-C to total carbon and water-soluble carbon in smoke PM 2.5 emitted from BB are 8.0-21.7 and 56.9-66.1 %, respectively. The corresponding contributions in smoke PM 2.5 from coal combustion are 5.2 and 45.5 %, respectively. These results suggest that BB and coal combustion are both important sources of HULIS in atmospheric aerosols. However, HULIS in diesel soot only accounted for ∼ 0.8 % of the soot particles, suggesting that vehicular exhaust may not be a significant primary source of HULIS. Primary HULIS and atmospheric HULIS display many similar chemical characteristics, as indicated by the instrumental analytical characterization, while some distinct features were also apparent. A high spectral absorbance in the UV-vis spectra, a distinct band at λ ex /λ em ≈ 280/350 nm in EEM spectra, lower H / C and O / C molar ratios, and a high content of [Ar-H] were observed for primary HULIS. These results suggest that primary HULIS contain more aromatic structures, and have a lower content of aliphatic and oxygen-containing groups than atmospheric HULIS. Among the four primary sources of HULIS, HULIS from BB had the highest O / C molar ratios (0.43-0.54) and [H-C-O] content (10-19 %), indicating that HULIS from this source mainly consisted of carbohydrate-and phenolic-like structures. HULIS from coal combustion had a lower O / C molar ratio (0.27) and a higher content of [Ar-H] (31 %), suggesting that aromatic compounds were extremely abundant in HULIS from this source. Moreover, the absorption Ångström exponents of primary HULIS from BB and coal combustion were 6.7-8.2 and 13.6, respectively. The mass absorption efficiencies of primary HULIS from BB and coal combustion at 365 nm (MAE 365 ) were 0.97-2.09 and 0.63 m 2 gC −1 , respectively. Noticeably higher MAE 365 values for primary HULIS from BB than coal combustion indicate that the former has a stronger contribution to the light-absorbing properties of aerosols in the atmospheric environment.

Quantifying primary and secondary humic-like substances in urban aerosol based on emission source characterization and a source-oriented air quality model

“…The percentages of HULIS C in WSOC for summer, autumn, winter, and spring, respectively, were 66.7 % ± 5.4 %, 54.1 % ± 11.2 %, 62.3 % ± 5.7 %, and 56.6 % ± 6.3 %, with an annual average of 59.5 % ± 9.2 %, suggesting that HULIS C was the major constituent of WSOC. This value is comparable to the results (about 60 %) at urban sites in the PRD region (Lin et al, 2010b;Fan et al, 2012), Shanghai (Zhao et al, 2015), South Korea (Park et al, 2012), Budapest (Salma et al, 2007(Salma et al, , 2010, and high-alpine areas of Jungfraujoch, Switzerland (Krivácsy et al, 2001). However, it is higher than the rural areas in K-puszta, Hungary (Salma et al, 2010), and the northeastern US (Pavlovic and Hopke, 2012).…”

“…HULIS isolation was based on the extraction method developed by Varga et al (2001) and used in many other studies (Nguyen et al, 2014;Lin et al, 2010b;Lin and Yu, 2011;Fan et al, 2012;Song et al, 2012;Salma et al, 2013;Feczko et al, 2007;Krivácsy et al, 2008). The separation procedure is provided in Text S1 of the Supplement.…”

Structural changes of CAST soot during a thermal–optical measurement protocol

“…Ferrari and Robertson (2000) propose an increase in L a with increasing I D /I G below L a ≈ 2 nm. In this regime I D /I G is proportional to the probability to find 6-fold aromatic rings in the cluster (Ferrari and Robertson, 2000). For larger crystallite sizes I D /I G decreases with increase in L a (Tuinstra and Koenig, 1970).…”

“…Raman scattering has been used for the structural investigation of soot for several years (Rosen and Novakov, 1978;Sadezky et al, 2005;Ferrari and Robertson, 2000;Schmid et al, 2011;Knauer et al, 2009;Ivleva et al, 2007a, b). It is sensitive to different C-C bonding types (e.g., graphitic structures) in a material (Sadezky et al, 2005) and depends on the ordering of sp 2 sites (Ferrari and Robertson, 2000).…”

“…The ratio of the intensities of the D and G peak (I D /I G ) can be linked to the crystallite size (L a ) within the investigated material (Ferrari and Robertson, 2000;Tuinstra and Koenig, 1970). Ferrari and Robertson (2000) propose an increase in L a with increasing I D /I G below L a ≈ 2 nm. In this regime I D /I G is proportional to the probability to find 6-fold aromatic rings in the cluster (Ferrari and Robertson, 2000).…”