Elsevier

Atmospheric Environment

Volume 73, July 2013, Pages 62-72
Atmospheric Environment

Size distribution of water-soluble components in particulate matter emitted from biomass burning

https://doi.org/10.1016/j.atmosenv.2013.03.025Get rights and content

Highlights

  • Size-resolved measurements of particulate matter (PM) emissions from biomass burning.

  • Unimodal size distributions were found for PM and its chemical species.

  • Relative composition of size-resolved biomass smoke is fairly constant.

  • Cl/K+ and SO42−/K+ varied strongly with particle sizes and biomass types.

Abstract

Size-resolved measurements of particulate matter (PM) emissions from 10 biomass materials (rice straw, soybean stem, green perilla stem, red pepper stem, pine needles, cherry leaves, cherry stem, maple leaves, gingko leaves and gingko stem) were conducted in a laboratory hood chamber environment using a 10-stage MOUDI. Samples were analyzed to determine the mass, water soluble organic carbon (WSOC), and water soluble inorganic species. This study examines how particle emissions and size distributions of chemical components vary with biomass materials. Mass fractions of water soluble organic mass (WSOM) (=1.6 × WSOC) and ionic species to the PM1.8 emissions varied significantly depending on the biomass type burned. The percent mass of WSOM in PM1.8 emissions ranged from 19.8% (green perilla stem) to 41.9% (red pepper stem) for agricultural crop residues, while the tree category accounted for 9.6% (gingko leaves) to 44.0% (gingko stem) of the PM1.8 emissions. Total ionic species contents in the PM1.8 mass ranged from 7.4% (rice straw) to 26.9% (green perilla stem) for the agricultural waste category, and 5.8% (maple leaves) to 23.5% (gingko stem) for the tree category. The ionic species fraction of the PM1.8 emission was dominated by K+, Cl, and SO42, while Ca2+ was important in the coarse mode particles (>3.1 μm). PM1.8 emissions of K+, Cl, and SO42 were as high as 16.9%, 9.0%, and 5.8%, respectively, and were from the green perilla stem, red pepper stem, and gingko stem emissions.

Normalized size distributions of mass, WSOC, K+, Cl, SO42, and oxalate in the biomass burning emissions showed a unimodal size distribution, peaking in the size ranges of 0.32–0.55 μm and 0.55–1.0 μm. Size-resolved PM mass fractions of WSOM, K+, Cl, and SO42 showed fairly consistent distributions for each biomass type, with higher fractions in the ultrafine mode (<0.10 μm) and lower fractions in the accumulation mode of 0.32–1.0 μm. The size distributions of WSOC were strongly correlated (mostly R2 > 0.90) with those of K+ in the particle size range of <0.1 μm and 0.1–1.8 μm and the biomass types. Strong correlations between the concentrations of K+–Cl and K+SO42 were observed for the following size ranges; <0.1 μm, 0.1–1.0 μm, 1.0–1.8 μm, and 1.8–3.1 μm for most biomass burning emissions. Regression line slopes for K+/WSOC (i.e., mass of K+/mass of WSOC from biomass burning emissions) were not significantly changed for particle size and biomass type, but slopes for Cl/K+ and SO42/K+ varied significantly with the particle size (ultrafine, condensation, droplet, and coarse modes) and biomass type.

Introduction

Biomass burning is as an important source of atmospheric particulate matter (PM) emissions around the world (Andreae and Merlet, 2001; Streets et al., 2003; Bond et al., 2004; Bo et al., 2008). Variation of burning material, terrain, burning type, combustion efficiency, and weather results in different chemical compositions and size distributions of particles emitted from biomass burning (Allen and Miguel, 1995; Turn et al., 1997; McDonald et al., 2000; Artaxo et al., 2002; Rissler et al., 2006; Fuzzi et al., 2007; Frey et al., 2009; Kim Oanh et al., 2011). Ambient concentrations of primary aerosol emissions from combustion sources are more elevated near sources than is usually the case for long range transported air. Aged aerosols tend to contain larger particles (>0.3 μm) than fresh primary aerosols from local combustion sources (Ondov and Wexler, 1998). Hence particle size distribution data could provide important information on source proximity. Their unique compositions and particle size provide the basis for receptor modeling and, thus, for tracing their transport in the atmosphere. Therefore, it is important to determine both the chemical composition and size distribution of the source to estimate the quantitative contribution of the emission source to a receptor site.

Size distribution measurements of PM emissions from biomass burning have typically been conducted in ambient air environments (Huang et al., 2006; Yang et al., 2006; Lee et al., 2008; Timonen et al., 2008; Saarnio et al., 2010). Mass size distribution of primary particle emissions from high temperature combustion sources was narrow with maximum in the accumulation mode size range with geometric mean diameters between 0.1 and 0.3 μm (Ondov and Wexler, 1998). Other studies have shown a unimodal size distribution of PM from biomass burning with a peak at 0.3–0.6 μm (Hays et al., 2005; Timonen et al., 2008; Saarnio et al., 2010). The size distribution of water-soluble organic carbon (WSOC) and inorganic ions in fresh smoke plumes from boreal wild fires in Europe indicated that the dominant mode for WSOC was typically in the size range of 0.1–1.0 μm (Saarnio et al., 2010). Normalized size distribution of long-ranged transported WSOC from wild fires also peaked in the range of 0.32–0.56 μm (Timonen et al., 2008). Kim Oanh et al. (2011) indicated that the largest fractions of PM, OC and elemental carbon emitted during the in-situ rice straw field burning experiments were associated with particle sizes of 0.4–1.1 μm. The size distributions of mass and chemical species in PM emissions from biomass burning in laboratory chamber environments can be used to understand changes in behavior and physicochemical characteristics of the smoke particles during transport. The consistency of the relative composition of size-resolved biomass smoke is important for source apportionments of ultrafine particles, PM0.2, PM0.5, etc. (Kleeman et al., 2008). To the best of our knowledge, size-resolved measurements of PM from biomass burning emissions in laboratory environments are extremely limited (Hays et al., 2005; Sippula et al., 2007; Kleeman et al., 2008; Frey et al., 2009), and it is not completely understood how mass size distributions of PM and its chemical species, such as WSOC, oxalate, Cl, SO42, and K+, vary with burned biomass materials. In addition, there is very little data in the literature addressing the variability of PM composition for biomass type and particle size.

In this study, size-resolved measurements of PM emissions from 10 biomass materials were made using a 10-stage MOUDI to determine mass, water-soluble organic and inorganic components. PM chemical abundance and size distribution results are presented and compared with those obtained from previous studies. The variability of the relative biomass smoke composition linked with biomass type and particle size are addressed.

Section snippets

Sampling of particulate matter emissions from biomass burning

In this study a total of 12 biomass materials were combusted to examine characteristics of particle emissions and size distributions of chemical components. The materials burned included four agricultural crop residues (rice straw, soybean stem, green perilla stem, and red pepper stem) and eight forest tree types (pine needles, pine stem, cherry leaves, cherry stem, maple leaves, maple stem, gingko stem, and gingko leaves). In Korea, crop residues are typically open burned in agricultural

Characteristic of water-soluble organic carbon in PM1.8

Only 10 of the 12 biomass combustion experiments are discussed due to measurement and/or analytical errors. Rice straw, soybean stem, green perilla stem, and red pepper stem are classified as agricultural crop residues, while pine needles, cherry leaves, cherry stem, maple leaves, gingko stem, and gingko leaves are classified as forest trees. The fraction of water-soluble components in the PM1.8 emission calculated from the sum of MOUDI impactor stages up to PM1.8 for each biomass type is

Conclusions

Size-resolved measurements of particulate matter emissions from combustion of 10 biomass materials were made to determine their gravimetric mass and water-soluble organic and inorganic components. Variances in emissions from different biomass burning sources and the observed differences in WSOC, K+, Cl, and SO42 concentrations among these sources were examined. Mass fractions of WSOM, K+, Cl, and SO42 in PM1.8 emissions were strongly dependent on the biomass type. Similar mass fraction

Acknowledgment

This study was financially supported by Chonnam National University, 2012. The authors would like to thank Kwang Yul Lee at GIST for his biomass burning sampling efforts.

References (30)

  • P. Artaxo et al.

    Physical and chemical properties of aerosols in the wet and dry seasons in Rondónia, Amazonia

    Journal of Geophysical Research

    (2002)
  • Y. Bo et al.

    Spatial and temporal variation of historical anthropogenic NMVOCs emission inventories in China

    Atmospheric Chemistry and Physics

    (2008)
  • T.C. Bond et al.

    A technology-based global inventory of black and organic carbon emissions from combustion

    Journal of Geophysical Research

    (2004)
  • A.K. Frey et al.

    Chemical composition and mass size distribution of fine particulate matter emitted by a small masonry heater

    Boreal Environment Research

    (2009)
  • S. Fuzzi et al.

    Overview of the inorganic and organic composition of size-segregated aerosol in Rondónia, Brazil, from the biomass-burning period to the onset of the wet season

    Journal of Geophysical Research

    (2007)
  • Cited by (69)

    • Variations in traffic-related water-soluble inorganic ions in PM<inf>2.5</inf> in Kanazawa, Japan, after the implementation of a new vehicle emission regulation

      2021, Atmospheric Pollution Research
      Citation Excerpt :

      It is worth noting that the [NO3−]/[SO42−] ratio in this study is close to the average level (0.12 ± 0.12) of the ratio derived from coal combustion (Bhanarkar et al., 2008; Chow et al., 2004; Guo et al., 2006; Hu et al., 2015; Liu et al., 2019; Matawle et al., 2014; Wei et al., 2019; Yan et al., 2020; Yang et al., 2021a; Zhang et al., 2012). In addition, the average value of the [NO3−]/[SO42−] ratio derived from biomass combustion in previous studies was 0.52 ± 0.48 (Allen and Miguel, 1995; Fine et al., 2001; Gao et al., 2008; Hays et al., 2005; Park et al., 2013; Schmidl et al., 2008a, 2008b; Sillapapiromsuk et al., 2019; Sun et al., 2019; Wang et al., 2007), which overlaps with the ratio range of coal combustion and traffic emissions. Therefore, we suggest that the [NO3−]/[SO42−] ratio should be used with caution to distinguish traffic sources and stationary sources in Japan and other countries that have implemented traffic emission regulations.

    View all citing articles on Scopus
    View full text