Five primary sources of organic aerosols in the urban atmosphere of Belgrade (Serbia)

Highlights

• Belgrade aerosols are affected by five primary sources.

• The increase of the levoglucosan/mannosan ratio characterizes the open fire period.

• The enantiomeric ratio (D/L) value for amino acids is higher in open combustions.

• Tracers from wood and other matrices involved in combustions help discerning events.

Abstract

Biomass burning and primary biological aerosol particles (PBAPs) represent important primary sources of organic compounds in the atmosphere. These particles and compounds are able to affect climate and human health. In the present work, using HPLC-orbitrapMS, we determined the atmospheric concentrations of molecular markers such as anhydrosugars and phenolic compounds that are specific for biomass burning, as well as the concentrations of sugars, alcohol sugars and d- and l-amino acids (D-AAs and L-AAs) for studying PBAPs in Belgrade (Serbia) aerosols collected in September–December 2008. In these samples, high levels of all these biomarkers were observed in October. Relative percentages of vanillic (V), syringic compounds (S) and p-coumaric acid (PA), as well as levoglucosan/mannosan (L/M) ratios, helped us discriminate between open fire events and wood combustion for domestic heating during the winter. L-AAs and D-AAs (1% of the total) were observed in Belgrade aerosols mainly in September–October. During open fire events, mean D-AA/L-AA (D/L) ratio values of aspartic acid, threonine, phenylalanine, alanine were significantly higher than mean D/L values of samples unaffected by open fire. High levels of AAs were observed for open biomass burning events.

Thanks to four different statistical approaches, we demonstrated that Belgrade aerosols are affected by five sources: a natural source, a source related to fungi spores and degraded material and three other sources linked to biomass burning: biomass combustion in open fields, the combustion of grass and agricultural waste and the combustion of biomass in stoves and industrial plants. The approach employed in this work, involving the determination of specific organic tracers and statistical analysis, proved useful to discriminate among different types of biomass burning events.

Introduction

Primary sources emit water-soluble organic compounds directly into the atmosphere; the contributions of biomass burning and biological material are among the most relevant ones.

Biomass burning represents the largest source of primary fine carbonaceous matter, the second most important source of gases (Akagi et al., 2011, Calvo et al., 2013) and an important emitter of toxic compounds in the atmosphere (Lemieux et al., 2004, Naeher et al., 2007, Torres-Duque et al., 2008). In addition, biomass burning aerosols influence the climate by affecting the solar balance of the Earth (Hobbs et al., 1997, Carmichael et al., 2009), by influencing snow albedo (Stocker et al., 2013, Flanner et al., 2007, Ramanathan and Carmichael, 2008), and by acting as cloud condensation nuclei (Novakov and Corrigan, 1996, Vestin et al., 2007).

Biomass burning includes both natural and anthropogenic (Akagi et al., 2011, Calvo et al., 2013) sources: open fires in forests, savannas, deforestation, shifting cultivation, and burning of agricultural waste. Although volcanic activities and lightning may cause the ignition of forest fires, humans are generally the starting source of ignition, whether deliberately or accidentally (Taylor, 2010). Moreover, biomass represents an energy source exploited by humans for domestic heating and cooking.

In the atmosphere, sources of primary biological aerosol particles (PBAPs) include biological materials such as microorganisms, fungal spores, bacteria, viruses, pollen, dispersal units, fragments of plants, cells and animals, and excretions from biological organisms (Despres et al., 2012). PBAPs are ubiquitous and are emitted directly into the atmosphere. Bioaerosol can contribute 20–30% of the total atmospheric particulate matter (> 0.2 μm) (Chen et al., 2013) and, in the Amazonian region, it can represent 85% of the mass of coarse particles (Poeschl et al., 2010).

PBAPs influence the climatic system by acting as cloud condensation nuclei and as ice nuclei. There are also indications of their importance in the cloud water cycle (Manninen et al., 2014) (and references therein). Bioaersols are aerodynamically buoyant and can be transported over long distances (Despres et al., 2012), thereby contributing to the diffusion of trace elements and pathogens far from their original location (Yang et al., 2012) (and references therein).

PBAPs also affect humans, animals and vegetation. Pollen and fungal spores are responsible for allergies in humans (Manninen et al., 2014), especially in the spring. The composition and abundance of PBAPs in the atmosphere are related to season, meteorological conditions (temperature and humidity are the most important factors), geographical location and human activities (Manninen et al., 2014, Yang et al., 2012) (and references therein).

In the present work, we determined the atmospheric concentration of organic tracers that are specific of biomass burning: phenolic compounds (syringic acid (SyA), isovanillic acid, homovanillic acid (HA), p-coumaric acid (PA), coniferyl aldehyde (CAH), vanillic acid (VA), vanillin (VAN), syringaldehyde (SyAH) and ferulic acid (FA)) (Simoneit, 2002) and anhydrosugars (levoglucosan, mannosan and galactosan). We also determined other compounds related to PBAPs as sugars (arabinose, mannose, xylose, galactose, glucose, fructose and sucrose), alcohol sugars (xylitol, arabitol, ribitol, sorbitol and galactitol, mannitol, glycerol, erythritol, maltitol) (Medeiros et al., 2006, Simoneit et al., 2004) and D- and L-AAs (L-Ala, L-Asp, L-Asn, L-Arg, L-Glu, L-Phe, L-Pro, L-Tyr, L-Thr, L-Hys, L-Lys, L-Leu/Ile, L-Orn, L-Ser, L-Gln, L-Val, Gly, D-Ala, D-Asp, D-Phe, D-Ser, and D-Thr) (Ge et al., 2011). Although biomass combustion often produces compounds such as sugars and alcoholsugars, these are also related to PBPAs (Jia et al., 2010b) (and references therein).

Phenolic compounds (PCs) are produced during biomass burning processes based on lignin combustion (Simoneit, 2002). Lignin is a biopolymer that is composed of three different aromatic alcohols: p-coumaryl, coniferyl and sinapyl alcohols. Their proportions differ among the major plant classes. The degradation products from the oxidation or burning of lignin are classified as coumaryl, vanillyl and syringyl moieties (Simoneit, 2002). In atmospheric aerosols, PCs may indicate the types of burned plants. Softwoods (gymnosperms) (Oros and Simoneit, 2001a) contain high proportions of coniferyl alcohol and minor proportions of sinapyl alcohol, and wood combustion produces primarily vanillyl moieties. The dominant PCs produced are vanillin, homovanillic acid, vanillic acid, and homovanillyl alcohol. Hardwood (angiosperm) lignin (Oros and Simoneit, 2001b) is enriched in sinapyl alcohol precursors, the combustion of these plants principally produces syringyl and vanillyl moieties. In deciduous tree smoke, the main methoxy phenols produced include homovanillyl alcohol, vanillic acid, vanillin, and syringic acid. In grasses (gramineae) (Oros et al., 2006), p-coumaryl alcohol is the prevalent lignin unit. Other significant products from burning grasses are acetosyringone, syringic acid, vanillin and vanillic acid.

Sugars can derive from numerous sources. Cellulosic material combustion produces anhydrosugars, which are specific markers of biomass burning (Simoneit, 2002) in the atmosphere. The combustion of hemicellulose generates high quantities of levoglucosan (from cellulose) and lower amounts of mannosan and galactosan.

Primary saccharides are produced from microorganisms, plants, animals, lichens, and bacteria (Dahlman et al., 2003, Simoneit et al., 2004, Yttri et al., 2007), while alcohol sugars come from fungal spores, bacteria, and lower plants (Bauer et al., 2008, Medeiros et al., 2006). Biomass burning is a source of saccharides (Di Filippo et al., 2013, Medeiros and Simoneit, 2008, Pio et al., 2008) due to the breakdown of polysaccharides and to the hydrolysis of anhydrosugars (Pio et al., 2008). It was also proposed as potential origin of fungal spores in urban areas. Sugars from soils and associated biota can be emitted in the atmosphere through re-suspension, erosion and agricultural activities (Jia et al., 2010a, Medeiros and Simoneit, 2007, Simoneit et al., 2004). Glucose, fructose and sucrose may derive from plant pollen and developing leaves (Fu et al., 2012, Graham et al., 2003, Pashynska et al., 2002). Arabitol and mannitol are described as molecular markers for fungal spores (Bauer et al., 2008, Burshtein et al., 2011, Di Filippo et al., 2013, Elbert et al., 2007) reflecting the contribution of microbially degraded material during the leaf senescence period and the fungal reproduction season (Medeiros et al., 2006, Pashynska et al., 2002). Their concentrations can be enhanced by biomass burning (Di Filippo et al., 2013).

AAs are an important class of compounds originating from biological, terrestrial and marine organisms (Ge et al., 2011). In general, L-AAs are associated to animals, terrestrial plants and phytoplankton (Cowie and Hedges, 1992), and are used as primary production molecular markers. Gly is one of the most stable AAs (McGregor and Anastasio, 2001). In the atmosphere, Gly suggests a more aged aerosol (Samy et al., 2013) correlated to long-range transport (Barbaro et al., 2011, Samy et al., 2013). D-AAs are used as markers of bacterial material both in terrestrial and marine environments (bacterioplankton) (Friedman, 2010, Kaiser and Benner, 2008, McCarthy et al., 1998). Enantiomeric D/L ratios are used as indicators of bacterial origin. Indeed, peptidoglycan in microbial cellular walls contains a high proportion of D-Ala, D-Asp, D-Glu and D-Ser (Dittmar et al., 2001, McCarthy et al., 1998). Due to their low volatility, AAs have been observed in condensed phases such as particulate matter (Barbaro et al., 2011, Barbaro et al., 2015, Di Filippo et al., 2014, Matsumoto and Uematsu, 2005, Scalabrin et al., 2012), dew (Scheller, 2001), rain (Mace et al., 2003b, Mace et al., 2003c), fog (Zhang and Anastasio, 2003), microlayer (Kuznetsova et al., 2004, Matrai et al., 2008), lakes (Barbaro et al., 2014) and marine waters (Kuznetsova et al., 2004, Matrai et al., 2008, Sommerville and Preston, 2001).

In this study we developed two new analytical methods to determine anhydrosugars and PCs using HPLC-orbitrap MS. The aim of this work was to better understand how primary sources, such as biomass burning and PBAPs, affect the chemical composition of particulate matter during the seasonal transition from late summer to early winter in an urban environment. We performed this task by determining organic biomarkers. Moreover, we performed a source apportionment in order to assess the contribution of the studied primary sources and to highlight possible differences in the origins of the contributors to the particulate matter. We performed this study in Belgrade, an urban site where information on water-soluble organic compounds are scarce despite the importance of a source such as biomass burning (Glavonjic, 2011, Glavonjic and Oblak, 2012).