Environ Sci Pollut Res DOI 10.1007/s11356-014-3080-8
RESEARCH ARTICLE
Carbonyl compounds and dissolved organic carbon in rainwater of an urban atmosphere D. Balla & A. Papageorgiou & D. Voutsa
Received: 21 March 2014 / Accepted: 20 May 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract This study investigates the occurrence of carbonyl compounds in rainwater at the city of Thessaloniki, Northern Greece. The concentrations of carbonyl compounds (as sum of 14 compounds) ranged from 21.8 to 592 μg/L, mean concentration 119 μg/L. Formaldehyde, acetaldehyde, hexanal, glyoxal, and methylglyoxal were the dominant compounds. DOC concentrations in rainwater ranged from 0.46 to 21.3 mg/L. UV–Vis and fluorescence spectra characteristics showed variation among rain events. Carbonyl compounds were negatively correlated with temperature exhibited relatively higher concentrations in cold season. They also influenced by storm origin with higher concentrations under terrestrial air masses. Calm conditions enhance the concentrations of DOC. Wash out is an effective removal mechanism of DOC. Keywords Atmosphere . Aldehydes . DOC . Fluorescence . Formaldehyde . Precipitation . UV–Vis . Wash out
Introduction Carbonyl compounds are a subset of atmospheric VOCs that include aldehydes, ketones, and bicarbonyls. They are emitted in the atmosphere from primary sources or formed as secondary products through a number of photochemical reactions. Primary sources of carbonyl compounds in atmosphere include vehicular emissions, emissions from combustion and industrial processes, and natural vegetative processes. Carbonyl compounds are directly emitted from vehicles with Responsible editor: Gerhard Lammel D. Balla : A. Papageorgiou : D. Voutsa (*) Environmental Pollution Control Laboratory, Department of Chemistry, Aristotle University, 541 24 Thessaloniki, Greece e-mail: [email protected]
conventional fuels and biofuels. Several aldehydes have been identified in on-road emissions with formaldehyde (FA) and acetaldehyde (AA) being usually the most abundant (BanWeiss et al. 2008; Nelson et al. 2008; Tsai et al. 2012; Magara- Gomez et al. 2012; Cheng et al. 2014). Biomass burning and residential wood burning are also possible sources of these compounds, with acetaldehyde usually emitted at higher rates (Hall et al. 2012; Zhang et al. 2013, Vicente et al. 2011; Cerqueira et al. 2013). Moreover, carbonyl compounds are formed as photochemical oxidation products of biogenic (mainly in rural areas) and anthropogenic VOCs (Atkinson 2000; Luecken et al. 2012; Cheng et al. 2014). Due to the abovementioned sources, carbonyl compounds are usually detected in the ambient atmosphere (Matsumoto et al. 2005; Zhang et al. 2012; Villanueva et al. 2013; Kawamura et al. 2013; Cheng et al. 2014). Formaldehyde and acetaldehyde are the dominant compounds. Dialdehydes such as glyoxal (GL) and methylglyoxal (MG) also occurred in significant concentrations (Kawamura et al. 2013). Carbonyl compounds play an important role in the atmospheric chemistry since they are involved in various photochemical reactions and considered as precursors of secondary organic aerosols (Kirkland et al. 2013). Besides their significance in the atmospheric chemistry, there is also concern on these compounds due to their significant impacts on human health and ecosystem (EPA 1999; IARC 2006, 2012). Photolysis and reactions with OH radicals are the major removal processes of carbonyl compounds from the atmosphere (Atkinson 2000). Wet deposition is also considered as significant removal route that may have significant impact on aqueous chemistry and land vegetation (Matsumoto et al. 2005; Seyfioglu et al. 2006). Rainout and washout are the main removal mechanisms of atmospheric pollutants. Rainout includes processes that take place in clouds (i.e., nucleation, condensation, and gas dissolution), whereas washout is the process that scavenges air pollutants between clouds and earth
Environ Sci Pollut Res
surface (Seyfioglu et al. 2006). The concentration of carbonyl compounds in rainwater depends on their ambient gaseous concentrations, their water solubility and Henry’s constant, the frequency of rain events, and the rainfall intensity. This study aims at the investigation of carbonyl compounds in rainwater of an urban area. For this purpose, rainwater samples were collected at the area of Thessaloniki, northern Greece, and analyzed for 14 compounds. Moreover, the dissolved organic carbon and its optical characteristics, UV–vis absorbance, and fluorescence, were studied. The occurrence of carbonyl compounds and dissolved carbon in rainwater were discussed in relation to meteorological data, air mass origin, and gaseous atmospheric pollutants.
Experimental Area description Thessaloniki (40° 62′ E, 22° 95′ N) is the second largest city in Greece, hosting almost 1,200,000 inhabitants. It is a coastal city, situated at the head of Thermaikos Gulf (Fig. 1) being surrounded by hills to the north and the mountain of Hortiatis (1,200 m) to the east. Several industrial islands are located to the west-northwest. Thessaloniki experiences a rather Mediterranean climate with mean temperatures during winter and summer at 7 and 25.3 °C, respectively, and mean annual rainfall of 445 mm. The highest frequency of air pollution episodes is associated with the presence of anticyclonic conditions over northern Greece. Additionally, local atmospheric circulations such as sea-land breezes and drainage flows affect
Fig. 1 Map of the study area. Sampling site is shown
the development, evolution, and maintenance of air pollution levels. The city usually present increased concentrations of particulate matter, carbonaceous components, and secondary formed species (Samara et al. 2014a; Voutsa et al. 2014). Sample collection Rain samples were collected by an automatic wet/dry collector (Sierra Andersen). The collector was equipped with two round bottom HDPE buckets and a lid, which was controlled and moved by a rain sensor at the begging and the end of each rain event in order to avoid any contamination of rain samples from dry deposition. The collector was on the roof of a building on the university campus located at the city center. The sampling point is not directly influenced by the surrounding traffic emissions. Twenty-five rainwater samples were collected during the period August–March 2013. After the collection, the sample was immediately transferred to the laboratory, stored in amber glass vials with PTFE-lined screw caps, and kept at 3.0×105 3.7×103
P vapor pressure, kH Henry’s law constant
out by calculating the relative response factors based on the area of internal standard 1,2-dibromopropane. Five calibration standard solutions were used to produce calibration curves for the target compounds. The analytical standard contained the 14 carbonyl compounds purchased from Ultra Scientific. The standard solutions of carbonyl compounds at different concentrations were prepared in acetonitrile. Calibration curves obtained for each compound were linear (R2 >0.980). The method detection limits varied from 0.25 μg/L for butanal and 2.4 μg/L for formaldehyde. Precision in terms of relative standard deviation of seven replicates ranged from 6.7 for benzaldehyde to 18.6 % for hexanal. Accuracy in terms of mean recovery of seven spiked samples ranged from 85 % for glyoxal to 115 % for heptanal. Finally, the efficiency of derivatization as relative standard deviation of the relative response factor of the surrogate standard ranged from 3.5 to 19.9 %. Laboratory reagent blanks and field blanks were also examined by employing the same procedure. Dissolved organic carbon Dissolved organic carbon (DOC) was determined in rain samples filtered through 0.45 μm by a TOC-Vcsh analyzer (Shimadzu). All samples were analyzed in duplicate. The detection limit was 0.1 mg/L with relative standard deviation ≤5 %. UV–vis measurement UV–vis absorbance was measured in rain samples filtered through 0.45 μm by using a Hitachi U-2001 spectrophotometer. All samples were analyzed in duplicate.
Fluorescence measurement Fluorescence was measured in rain samples filtered through 0.45 μm by the F-7000 Fluorescence Spectrophotometer (Hitachi) equipped with a 150 W Xe arc lamp. Samples were allowed to equilibrate to room temperature before analyses. Synchronous fluorescence spectra was scanned from 240 to 400 nm, at Δλ=44 nm. The band pass of both excitation and emission monochromator was set at 5 nm. The scan speed was 240 nm/min, and the photomultiplier tube voltage was set at 700 V. Replicate scans were generally within 5 % agreement in terms of intensity and within band pass resolution. All samples were measured in duplicate. Air mass trajectories Five-day back trajectories were calculated using the HYSPLIT4 model for each rain event. The Hybrid Single Particle Lagranian Integrated Trajectory Model (HYSPLIT) (Draxler and Rolph 2013; Rolph 2013). A 72-h back trajectory analysis was performed for altitudes 500, 1,000, and 1,500 m. Statistical analysis The SPSS software (IBM SPSS, version 21) was used for statistical analysis of data. Concentrations below method detection limits were assigned a value equal to half of the method detection limit. Correlations between determined parameters were investigated by calculating Spearman’s correlation coefficients. Correlation coefficient significances at least at 95 % significance level are discussed. The differences among various groups were tested with the non-parametric Kruskal-Wallis rank test (p
RESEARCH ARTICLE
Carbonyl compounds and dissolved organic carbon in rainwater of an urban atmosphere D. Balla & A. Papageorgiou & D. Voutsa
Received: 21 March 2014 / Accepted: 20 May 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract This study investigates the occurrence of carbonyl compounds in rainwater at the city of Thessaloniki, Northern Greece. The concentrations of carbonyl compounds (as sum of 14 compounds) ranged from 21.8 to 592 μg/L, mean concentration 119 μg/L. Formaldehyde, acetaldehyde, hexanal, glyoxal, and methylglyoxal were the dominant compounds. DOC concentrations in rainwater ranged from 0.46 to 21.3 mg/L. UV–Vis and fluorescence spectra characteristics showed variation among rain events. Carbonyl compounds were negatively correlated with temperature exhibited relatively higher concentrations in cold season. They also influenced by storm origin with higher concentrations under terrestrial air masses. Calm conditions enhance the concentrations of DOC. Wash out is an effective removal mechanism of DOC. Keywords Atmosphere . Aldehydes . DOC . Fluorescence . Formaldehyde . Precipitation . UV–Vis . Wash out
Introduction Carbonyl compounds are a subset of atmospheric VOCs that include aldehydes, ketones, and bicarbonyls. They are emitted in the atmosphere from primary sources or formed as secondary products through a number of photochemical reactions. Primary sources of carbonyl compounds in atmosphere include vehicular emissions, emissions from combustion and industrial processes, and natural vegetative processes. Carbonyl compounds are directly emitted from vehicles with Responsible editor: Gerhard Lammel D. Balla : A. Papageorgiou : D. Voutsa (*) Environmental Pollution Control Laboratory, Department of Chemistry, Aristotle University, 541 24 Thessaloniki, Greece e-mail: [email protected]
conventional fuels and biofuels. Several aldehydes have been identified in on-road emissions with formaldehyde (FA) and acetaldehyde (AA) being usually the most abundant (BanWeiss et al. 2008; Nelson et al. 2008; Tsai et al. 2012; Magara- Gomez et al. 2012; Cheng et al. 2014). Biomass burning and residential wood burning are also possible sources of these compounds, with acetaldehyde usually emitted at higher rates (Hall et al. 2012; Zhang et al. 2013, Vicente et al. 2011; Cerqueira et al. 2013). Moreover, carbonyl compounds are formed as photochemical oxidation products of biogenic (mainly in rural areas) and anthropogenic VOCs (Atkinson 2000; Luecken et al. 2012; Cheng et al. 2014). Due to the abovementioned sources, carbonyl compounds are usually detected in the ambient atmosphere (Matsumoto et al. 2005; Zhang et al. 2012; Villanueva et al. 2013; Kawamura et al. 2013; Cheng et al. 2014). Formaldehyde and acetaldehyde are the dominant compounds. Dialdehydes such as glyoxal (GL) and methylglyoxal (MG) also occurred in significant concentrations (Kawamura et al. 2013). Carbonyl compounds play an important role in the atmospheric chemistry since they are involved in various photochemical reactions and considered as precursors of secondary organic aerosols (Kirkland et al. 2013). Besides their significance in the atmospheric chemistry, there is also concern on these compounds due to their significant impacts on human health and ecosystem (EPA 1999; IARC 2006, 2012). Photolysis and reactions with OH radicals are the major removal processes of carbonyl compounds from the atmosphere (Atkinson 2000). Wet deposition is also considered as significant removal route that may have significant impact on aqueous chemistry and land vegetation (Matsumoto et al. 2005; Seyfioglu et al. 2006). Rainout and washout are the main removal mechanisms of atmospheric pollutants. Rainout includes processes that take place in clouds (i.e., nucleation, condensation, and gas dissolution), whereas washout is the process that scavenges air pollutants between clouds and earth
Environ Sci Pollut Res
surface (Seyfioglu et al. 2006). The concentration of carbonyl compounds in rainwater depends on their ambient gaseous concentrations, their water solubility and Henry’s constant, the frequency of rain events, and the rainfall intensity. This study aims at the investigation of carbonyl compounds in rainwater of an urban area. For this purpose, rainwater samples were collected at the area of Thessaloniki, northern Greece, and analyzed for 14 compounds. Moreover, the dissolved organic carbon and its optical characteristics, UV–vis absorbance, and fluorescence, were studied. The occurrence of carbonyl compounds and dissolved carbon in rainwater were discussed in relation to meteorological data, air mass origin, and gaseous atmospheric pollutants.
Experimental Area description Thessaloniki (40° 62′ E, 22° 95′ N) is the second largest city in Greece, hosting almost 1,200,000 inhabitants. It is a coastal city, situated at the head of Thermaikos Gulf (Fig. 1) being surrounded by hills to the north and the mountain of Hortiatis (1,200 m) to the east. Several industrial islands are located to the west-northwest. Thessaloniki experiences a rather Mediterranean climate with mean temperatures during winter and summer at 7 and 25.3 °C, respectively, and mean annual rainfall of 445 mm. The highest frequency of air pollution episodes is associated with the presence of anticyclonic conditions over northern Greece. Additionally, local atmospheric circulations such as sea-land breezes and drainage flows affect
Fig. 1 Map of the study area. Sampling site is shown
the development, evolution, and maintenance of air pollution levels. The city usually present increased concentrations of particulate matter, carbonaceous components, and secondary formed species (Samara et al. 2014a; Voutsa et al. 2014). Sample collection Rain samples were collected by an automatic wet/dry collector (Sierra Andersen). The collector was equipped with two round bottom HDPE buckets and a lid, which was controlled and moved by a rain sensor at the begging and the end of each rain event in order to avoid any contamination of rain samples from dry deposition. The collector was on the roof of a building on the university campus located at the city center. The sampling point is not directly influenced by the surrounding traffic emissions. Twenty-five rainwater samples were collected during the period August–March 2013. After the collection, the sample was immediately transferred to the laboratory, stored in amber glass vials with PTFE-lined screw caps, and kept at 3.0×105 3.7×103
P vapor pressure, kH Henry’s law constant
out by calculating the relative response factors based on the area of internal standard 1,2-dibromopropane. Five calibration standard solutions were used to produce calibration curves for the target compounds. The analytical standard contained the 14 carbonyl compounds purchased from Ultra Scientific. The standard solutions of carbonyl compounds at different concentrations were prepared in acetonitrile. Calibration curves obtained for each compound were linear (R2 >0.980). The method detection limits varied from 0.25 μg/L for butanal and 2.4 μg/L for formaldehyde. Precision in terms of relative standard deviation of seven replicates ranged from 6.7 for benzaldehyde to 18.6 % for hexanal. Accuracy in terms of mean recovery of seven spiked samples ranged from 85 % for glyoxal to 115 % for heptanal. Finally, the efficiency of derivatization as relative standard deviation of the relative response factor of the surrogate standard ranged from 3.5 to 19.9 %. Laboratory reagent blanks and field blanks were also examined by employing the same procedure. Dissolved organic carbon Dissolved organic carbon (DOC) was determined in rain samples filtered through 0.45 μm by a TOC-Vcsh analyzer (Shimadzu). All samples were analyzed in duplicate. The detection limit was 0.1 mg/L with relative standard deviation ≤5 %. UV–vis measurement UV–vis absorbance was measured in rain samples filtered through 0.45 μm by using a Hitachi U-2001 spectrophotometer. All samples were analyzed in duplicate.
Fluorescence measurement Fluorescence was measured in rain samples filtered through 0.45 μm by the F-7000 Fluorescence Spectrophotometer (Hitachi) equipped with a 150 W Xe arc lamp. Samples were allowed to equilibrate to room temperature before analyses. Synchronous fluorescence spectra was scanned from 240 to 400 nm, at Δλ=44 nm. The band pass of both excitation and emission monochromator was set at 5 nm. The scan speed was 240 nm/min, and the photomultiplier tube voltage was set at 700 V. Replicate scans were generally within 5 % agreement in terms of intensity and within band pass resolution. All samples were measured in duplicate. Air mass trajectories Five-day back trajectories were calculated using the HYSPLIT4 model for each rain event. The Hybrid Single Particle Lagranian Integrated Trajectory Model (HYSPLIT) (Draxler and Rolph 2013; Rolph 2013). A 72-h back trajectory analysis was performed for altitudes 500, 1,000, and 1,500 m. Statistical analysis The SPSS software (IBM SPSS, version 21) was used for statistical analysis of data. Concentrations below method detection limits were assigned a value equal to half of the method detection limit. Correlations between determined parameters were investigated by calculating Spearman’s correlation coefficients. Correlation coefficient significances at least at 95 % significance level are discussed. The differences among various groups were tested with the non-parametric Kruskal-Wallis rank test (p
