Atmospheric Environment 84 (2014) 172e177
Contents lists available at ScienceDirect
Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv
Variability of ethanol and acetaldehyde concentrations in rainwater R.J. Kieber*, S. Tatum, J.D. Willey, G.B. Avery, R.N. Mead Department of Chemistry and Biochemistry, University of North Carolina Wilmington, Wilmington, NC 28403-5932, United States
h i g h l i g h t s Average concentration of ethanol and acetaldehyde in rain are 192 nM and 193 nM. Ethanol and acetaldehyde variability driven by temporal and air mass back trajectory. Ethanol results represent baseline for concentrations in North American rainwater.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 June 2013 Received in revised form 13 November 2013 Accepted 14 November 2013
Ethanol and acetaldehyde concentrations were measured in 52 rain events collected between January 25, 2011 and March 4, 2012 in Wilmington, North Carolina, USA. Ethanol concentrations ranged from 23 nM to 908 nM with a volume weighted average concentration of 192 20 nM while acetaldehyde ranged from 23 nM to 909 nM with a volume weighted average concentration of 193 25 nM. There was a great deal of variability in the abundance of ethanol and acetaldehyde between rain events driven primarily by temporal and air mass back trajectory influences. The ratio of ethanol to acetaldehyde was at a minimum during periods of peak solar intensity underscoring the importance of alcohol oxidation by a photochemically generated oxidant such as hydroxyl radical in the gas and/or aqueous phase. Ethanol and acetaldehyde concentrations were not strongly correlated with rain amount suggesting that gas-phase concentrations were not significantly depleted during the storm or that they were resupplied during the course of the rain event. The concentration of ethanol and acetaldehyde were correlated with nitrate and non-sea salt sulfate suggesting the importance of terrestrial and anthropogenic inputs at this location. Comparison of future ethanol and acetaldehyde concentrations in rainwater to the data presented in this study will help delineate potential consequences of these labile oxygenated volatile organic compounds (OVOCs) on the chemistry of the troposphere as the United States transitions to more ethanol blended fuels. Aqueous phase impacts of increasing ethanol concentrations will be particularly significant to the oxidizing capacity of atmospheric waters because of its reactivity with OH and HO2 radicals in solution. Increased rainwater concentrations could also have significant ramifications on receiving watersheds because of the biogeochemical lability of the alcohol. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Ethanol Acetaldehyde Rainwater
1. Introduction Ethanol is a chemically and biologically labile compound that has received a great deal of attention recently (Kirstine and Galbally, 2012; Naik et al., 2010) because of its dramatic increase in production and use as a biofuel both in the United States and abroad (de Gouw et al., 2012; Millet et al., 2012). Current estimates indicate that 10% of the United States gasoline supply is ethanol with more than 95% of gasoline sold containing added ethanol most commonly as E10 (de Gouw et al., 2012). The use of ethanol as a fuel will most likely increase significantly with the recent approval of
* Corresponding author. Tel.: þ1 910 962 3865; fax: þ1 910 962 3013. E-mail address:
[email protected] (R.J. Kieber). 1352-2310/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atmosenv.2013.11.038
E15 gasoline in light duty vehicles model year 2001 and newer by the United States Environmental Protection Agency (EPA). This upsurge in biofuel usage has significant ramifications because emission studies of vehicles utilizing ethanol-blended fuels demonstrate that significant quantities of ethanol are emitted uncombusted from tailpipes and that fuels with higher ethanol content emit higher levels of the alcohol (Poulopoulos et al., 2001). Enhanced vehicular ethanol emissions may impact a variety of important atmospheric processes including the oxidizing capacity of atmospheric waters because of its reactions with OH and HO2 radicals in solution (Naik et al., 2010 and references therein). Reactions of these oxidants with ethanol have also been linked to increases in ambient levels of acetaldehyde (Millet et al., 2012) that is a source of peroxyacetyl nitrate (PAN) and ozone (Naik et al., 2010 and references therein). Another potential unforeseen consequence
R.J. Kieber et al. / Atmospheric Environment 84 (2014) 172e177
to the increasing usage of ethanol is that it may play an important role in secondary aerosol formation (Blando and Turpin, 2000). Despite its documented reactivity in the troposphere, virtually nothing is known regarding the abundance of ethanol in atmospheric waters. Studies of ethanol concentrations in precipitation have been limited primarily by the inadequacy of existing analytical methods. Low molecular weight saturated straight chain alcohols (C1eC4) are difficult to quantify in aqueous environmental matrices because they are in very low concentrations, structurally similar to water, have poor molar absorptivities and are hard to derivatize for spectroscopic analysis. One limited study at Creteil University 15 km from Paris, France reported <1e5 mM ethanol measured in 7 discrete rain events by direct injection (Monod et al., 2003). The high analytical detection limit reported (1 mM) suggests that the data are useful primarily in comparison to highly urbanized locations but are not of adequate sensitivity to aid in the interpretation of ethanol biogeochemical cycling in precipitation collected from less urbanized locations where ethanol is produced and consumed. This study also lacked sufficient sampling frequency to allow for more detailed analysis of temporal or air mass back trajectory influences on ethanol concentrations. The overarching goal of the research presented here is to describe in detail the processes that influence and control ethanol distributions in rainwater. Specifically we define the ranges and patterns of variation in the abundance of rainwater ethanol including such factors as the influence of season and air mass back trajectory on concentrations. We also describe patterns of correlation between ethanol and other rainwater components such as dissolved organic carbon (DOC) and inorganic ions. The manuscript contains concurrent measurements of acetaldehyde which provide important mechanistic information regarding the cycling of ethanol in precipitation. 2. Experimental 2.1. Sample collection Wilmington rainwater samples were collected on an event basis on the campus of the University of North Carolina at Wilmington (UNCW) from January 2011 to March 2012. The collection site at UNCW is a large open area of approximately 1 ha and is made up of a turkey oak, long leaf pine and wire grass community. This area is typical of the inland coastal area of southeastern North Carolina. The site (3413.90 N, 77 52.70 W) is approximately 8.5 km from the Atlantic Ocean. Due to the close proximity of the collection site to the laboratory, ethanol analysis or filtration and refrigeration of samples can be done within minutes of collection, which reduces the possibility of compositional changes between the time of collection and analysis. Event rain samples were collected using Aerochem-Metrics (ACM) model 301 automatic sensing wet/dry precipitation collectors containing 4 L Pyrex glass beakers that were pre-cleaned by combusting at 450 C for 4 h to remove organic impurities. All reported samples were either collected and analyzed or filtered through a 0.2 mm polyesthersulphone filter using a Pyrex filtration apparatus and refrigerated in a 7 mL Teflon vials with within an hour after cessation of a rain event. Rainwater concentrations are reported as volume-weighted concentrations with volume-weighted standard deviations (Topol et al., 1985). This is the mathematical equivalent to mixing all rain within a specified time period together and reporting the analytical result for that composite sample. Precipitation events were categorized using air-mass back-trajectories generated using version 4 of the Hybrid Single Particle Lagrangian Integrated Trajectory Model (HYSPLIT) developed at the National Oceanic Atmospheric Administration Air Resources Laboratory (Draxler and Rolph, 2003). Trajectories were generated
173
using a web-based version of the model and calculated for each measured precipitation event collected at UNCW starting at the recorded end of precipitation for a 72 h hind-cast for altitudes of 500 and 1000 m. 2.2. Acetaldehyde and ethanol Acetaldehyde concentrations in rainwater samples were determined by derivatization with 2,4-dinitrophenylhydrazine followed by separation and detection by HPLC (Kieber et al., 1999). Samples and standards reacted with 2,4-dinitrophenylhydrazine for one hour in the dark forming a hydrazone, which was separated from interfering substances by HPLC and quantified by UV detection at 370 nm. Derivitized samples (100 mL) were injected onto a reversed phase Luna 100 mm 4.60 mm 3m C18 Phenomenex column with a 100 A pore size at 10 C. The mobile phase was a 1:1 mixture of 0.1% trifluoroacetic acid (TFA) in acetonitrile and 0.1% TFA in DIW at a flow rate of 1.00 mL min1. Ethanol was determined on a second aliquot by oxidation of the alcohol to acetaldehyde via alcohol oxidase obtained from the yeast Hansenula sp. (Kieber et al., 2013). The enzyme was prepared by dissolution of 100 units of alcohol oxidase in 5 mL of 0.1 M potassium phosphate buffer (pH 9.0). The sample (1000 mL) was combined with 10 mL of buffer, 100 mL of an enzyme working reagent (0.18 units mL1) and allowed to react at 40 C for 120 min before addition of 10 mL of DNPH. The concentration of ethanol was determined after HPLC analysis by the difference in acetaldehyde concentration in samples with and without added enzyme. This method has a detection limit of 10 nM and a precision of 2% RSD. Accuracy was verified via an intercomparison study of rainwater ethanol concentrations (n ¼ 26) utilizing an independent method employing solid phase micro extraction (SPME). Comparison of the resulting rainwater ethanol concentrations produced a trend line with a slope of unity with a 2% deviation between analytical results demonstrating that the methods produced statistically equivalent ethanol concentrations in precipitation samples (Kieber et al., 2013). 2.3. Supporting analyses Hydrogen peroxide was analyzed at the time of sample collection by a fluorescence decay technique involving the peroxidasemediated oxidation of the fluorophore scopoletin by H2O2 in rain buffered at a pH of 7 with a phosphate buffer (Mullaugh et al., 2011). Organic carbon content in rainwater samples were determined with a Shimadzu TOC 5000 carbon analyzer (Shimadzu, Kyoto, Japan) equipped with an ASI 5000 autosampler (Willey et al., 2 2000). Inorganic anions (Cl, NO 3 , and SO4 ) were analyzed using suppressed ion chromatography. A Ross electrode with low ionic strength buffers was used for pH analysis. These supporting data were used to characterize rain events and to evaluate whether the patterns of variation observed for ethanol co-vary with any of these analytes. These supporting data also allow comparison with rain collected elsewhere. 3. Results and discussion Ethanol and acetaldehyde concentrations were measured in 52 rain events collected between January 25, 2011 and March 4, 2012 in Wilmington, North Carolina. Ethanol concentrations ranged from 23 nM to 908 nM with a volume weighted average concentration of 192 20 nM while acetaldehyde ranged from 23 nM to 909 nM with a volume weighted average concentration of 193 25 nM. This sample set represents 34% of the rain events and 59% of the rain volume collected during this time period. Concentrations are reported as volume weighted averages (VWA) in order to decrease
174
R.J. Kieber et al. / Atmospheric Environment 84 (2014) 172e177
the influence of small rain events, which is equivalent to combining all samples and analyzing one representative sample. 3.1. Impact of season The ethanol and acetaldehyde concentrations were subdivided into 4 time periods in order to examine seasonal variations in analyte concentrations. Seasons were defined as winter (January 1e March 31), spring (April 1eJune 30), summer (July 1eSeptember 30), and fall (October 1eDecember 31). The volume weighted average concentration of ethanol was maximum in winter and spring and minimum during summer (t-test, p ¼ 0.020) suggesting a strong seasonal influence of ethanol concentrations in rainwater (Fig. 1). Observed seasonal concentrations of ethanol in rainwater most probably occur from a dynamic competition between various production and decay processes including production from biogenic and anthropogenic sources and oxidation by photochemically generated hydroxyl radicals. The high wintertime ethanol concentrations likely result from reduced sinks because the rate of photochemically generated hydroxyl radicals would be significantly smaller in the low irradiance winter months. The concentration of acetaldehyde was less than half that of ethanol while the ratio of ethanol to acetaldehyde (E:A) was at a maximum during winter (Table 1) consistent with the reduced role of photochemically generated oxidants in both the aqueous and gas phases during this time period. The large winter ethanol concentrations could also be driven to some extent by industrial emissions and vehicular exhaust of ethanol. An earlier study employing a box model demonstrated that combustion of E85 fuel emitted more ethanol at 7 C relative to 24 C suggesting that the higher winter values could result from increased low temperature E10 ethanol emissions (Ginnebaugh et al., 2010). The minimum in ethanol concentrations in summer is most likely the result of increased sinks rather than decreased source strengths. Increased solar radiation accompanied by high temperatures favors photo oxidation during summer by hydroxyl radical which is the primary sink of gas and aqueous phase ethanol (Naik et al., 2010). The summer minimum is most likely not due to solubility differences as the greatest change in ethanol concentration occurs between spring and summer where the temperature change of 4 C (Table 1) would correspond to a change in the Henry’s Law constant (KH) of only 6%(Warneck, 2006). The E:A ratio was also at a minimum during summer underscoring the importance of ethanol
Fig. 1. Volume weighted average concentration of ethanol and acetaldehyde grouped by season. Error bars represent volume weighted standard deviations. Number of events is under season name. Seasons were grouped as follows: winter ¼ January 1e March 31; spring ¼ April 1eJune 30; summer ¼ July 1eSeptember 30; fall ¼ October 1eDecember 31.
Table 1 Volume of rain (mm), ratio of ethanol to acetaldehyde (E:A) and average temperature ( C) during the four sampling seasons. The concentration of hydrogen peroxide is presented as a volume weighted average and volume weighted standard deviation during each season. Season
Amount (mm)
E:A
H2O2 (mM)
Winter Spring Summer Fall
246 199 156 125
2.43 0.67 0.44 1.75
10.1 25.4 20.2 5.6
2.0 3.0 3.3 1.6
T ( C) 8.9 22.6 26.6 14.6
oxidation by a photochemically generated oxidant during the high irradiance summer months (Table 1). 3.2. Dynamics of the ethanol:acetaldehyde ratio in rainwater Factors governing changes in ethanol and acetaldehyde concentrations in rainwater are most likely complex involving fluctuations in both source and sink strengths including such things as oscillations in biogenic emissions caused by photosynthetic activity as well as variations in boundary layer dynamics. Temporal changes in the ratio of ethanol to acetaldehyde (E:A) were examined because they provide important insight into the mechanistic underpinnings driving the cycling of the analytes in rainwater (Fig. 2). Each bar in Fig. 2 represents the ratio of ethanol to acetaldehyde concentrations during the given time period. Events were excluded from classification if they occurred during more than one time period. The E:A ratio exhibited a distinct oscillation with decreasing values throughout the day followed by a minimum in the afternoon (12 pm e 6 pm) during the period of peak solar intensity (Fig. 2) suggesting ethanol is oxidized by a photo-mediated process to acetaldehyde during daylight hours. Upon removal of sunlight the ratio reached its maximum value in the 6 pm 12 am time period suggesting ethanol is replenished by anthropogenic and/or biogenic emissions and is not as rapidly oxidized to acetaldehyde. 3.3. Influence of storm origin Rain events were classified as either marine or continental to determine how ethanol and acetaldehyde concentrations were affected by air mass back trajectory (Fig. 3). Other rain events that did not fit into one of these two end member storm classifications such as mixed stationary fronts and local thunderstorms were not included in Fig. 3. Back trajectories were generated at 500 m and 1000 m above ground level by NOAA HySplit v4. An additional storm classification was made with air mass back trajectories originating over the Cornbelt region of the United States where
Fig. 2. Ratio of ethanol to acetaldehyde (E:A) during four daily collection times periods.
R.J. Kieber et al. / Atmospheric Environment 84 (2014) 172e177
175
3.4. Correlation analysis
Fig. 3. Volume weighted average concentration of ethanol and acetaldehyde grouped by air mass back trajectory. Error bars represent VW standard deviations and the sample number represents the number of events in each trajectory.
most of the country’s ethanol biofuel is produced and consumed. This region consists of several states including Iowa, Illinois, Indiana, Michigan and parts of Nebraska, Kansas, Minnesota and Missouri. There was no significant difference between the concentration of ethanol in Cornbelt storms and other terrestrial storms (ANOVA, p ¼ 0.779) suggesting that transport from this region did not significantly influence ethanol concentrations relative to other continentally dominated rain events. Volume weighted ethanol and acetaldehyde concentrations were higher for storms of continental origin compared to marine events (ManneWhitney Rank Sum Test p ¼ 0.043 acetaldehyde; p ¼ 0.004 ethanol) similar to what has been observed for many other organic compounds at this location (Avery et al., 2006, 2013; Kieber et al., 2006; Willey et al., 2000) suggesting there is a significant continental source of both analytes at this location. The ratio of ethanol to acetaldehyde (E:A) was very similar in continentally dominated and marine events (1.3 vs. 1.4 respectively) suggesting that the processing of the alcohol to the aldehyde was not dramatically influenced by air mass back trajectory. The continental source for ethanol is most likely a combination of significant biogenic emissions and anthropogenic inputs including vehicular exhaust, the relative importance of which would vary by season (Kirstine and Galbally, 2012; Naik et al., 2010)(Fig. 1). Biogenic emissions of acetaldehyde as well as its precursor alkenes would also contribute to rainwater concentrations in terrestrial storms as well as inputs from vehicular exhaust (Millet et al., 2010). Much less is known regarding the source of ethanol and acetaldehyde in marine dominated rainwater relative to continental rainwater. Measurements of methanol and acetaldehyde in North Atlantic marine air at the Mace Head observatory suggest in situ atmospheric production as a source of OVOCs in clean marine air occurring via longer lived intermediates such as organic peroxides and long chain alcohols (Lewis et al., 2005). It is also possible there is direct in situ photochemical production of ethanol and acetaldehyde in the aqueous phase from the photodegradation of larger macromolecular species in marine rainwater similar to what has been reported for formaldehyde at this location (Southwell et al., 2010). Direct oceanic emissions of ethanol and acetaldehyde from photodegradation of chromophoric dissolved organic matter is another potential source of OVOCs in marine rain (Kieber et al., 1990). In a more recent study based on a limited number of oceanic measurements it was hypothesized that the ocean may be an important direct source of ethanol to the atmosphere especially in natural upwelling regions (Beale et al., 2010). It is likely a combination of these processes explain the relatively high ethanol and acetaldehyde concentrations in rainwater with reduced continental inputs.
The minimum in the E:A ratio and its correlation with peak solar radiation (Fig. 2) highlight the importance of ethanol oxidation by a photochemically generated oxidant such as hydroxyl radical in the gas phase, aqueous phase or both. It is difficult to measure the concentration of hydroxyl radical directly. However, the hydroxyl radical is produced in both the gas and aqueous phase by photolysis of hydrogen peroxide, which was measured in the same rainwater samples as ethanol and acetaldehyde. The concentration of H2O2was examined as a function of the ethanol to acetaldehyde (E:A) ratio in order to determine the influence of peroxide on the abundance of ethanol. There is a strong negative correlation (n ¼ 53; p < 0.01) between the E:A ratio and H2O2 for all rain events suggesting that when the concentration of peroxide is high significant oxidation of ethanol to acetaldehyde occurs. The most likely reason for this inverse correlation involves production of hydroxyl radicals generated from the photodegradation of H2O2. This reaction could occur in the gas phase and be subsequently propagated into the aqueous phase or the hydroxyl radical could be generated by peroxide photolysis in the aqueous phase followed by direct ethanol oxidation. The average seasonal ratio of ethanol to acetaldehyde (E:A) was also examined as a function of the volume weighted average hydrogen peroxide concentration measured during the analogous time period. The two highest VWA H2O2 concentrations and lowest E:A ratios occur during spring and summer while the highest E:A ratios occur during fall and winter corresponding to lower peroxide values (Table 1). A similar inverse relationship (n ¼ 4; p < 0.02) occurs when the E:A ratio is plotted as a function of hydrogen peroxide measured during different times of day (Fig. 4). This suggests the impact of photochemically produced hydrogen peroxide and hydroxyl radicals influence ethanol concentrations over short term daily cycles as well as seasonally. Ethanol and acetaldehyde concentrations were also plotted versus rain amount in order to investigate how the analytes varied as a function of sample volume (Fig. 5). Ethanol and acetaldehyde concentrations were not strongly correlated with rain amount suggesting that gas-phase concentrations were not significantly depleted during the storm or that they were resupplied during the course of the rain event. The highest volume storm (85 mm) had significant ethanol (158 nM) and acetaldehyde (450 nM) concentrations suggesting that local sources and/or in situ production are
Fig. 4. Ratio of ethanol to acetaldehyde (E:A) versus volume weighted average concentration of H2O2 during four daily collection times periods (n ¼ 4; p < 0.02).
176
R.J. Kieber et al. / Atmospheric Environment 84 (2014) 172e177
in order to examine how the strength of sources and sinks might impact the relationship between ethanol and acetaldehyde. As was the case when all data was combined, there was no correlation between the concentration of ethanol and acetaldehyde in any of the correlation matrices suggesting that the relationship between the abundance of the alcohol and aldehyde is complex in precipitation samples.
4. Implications
Fig. 5. Ethanol and acetaldehyde concentrations as a function of rain amount.
replenishing the atmosphere during the rain event. Formaldehyde also does not exhibit washout at this location (Kieber et al., 1999) which the authors attribute to in situ photochemical production from chromophoric dissolved organic matter in rainwater (Southwell et al., 2010). Pearson Correlation analyses were done to investigate relationships between ethanol, acetaldehyde and various other analytes measured in this study. Correlations with p < 0.05 were defined as significant. Concentrations of hydrogen ion, nitrate and non-sea salt sulfate (NSS) in rainwater were all intercorrelated in the current study as in many other previous studies at this location (Kieber et al., 2002, 2006, 2008). A strong correlation between Hþ, NO 3 and NSS is generally interpreted as reflecting the importance of anthropogenic sources hence they are often used as pollution indicators in rain. The concentration of ethanol and acetaldehyde were also correlated with nitrate and NSS (Table 2) suggesting the importance of anthropogenic inputs to rainwater at this location. The concentration of ethanol was not correlated to the abundance of dissolved organic carbon (DOC) suggesting that the alcohol makes up a small and variable fraction of the dissolved organic pool in precipitation. Surprisingly there is no correlation between the concentration of ethanol and acetaldehyde in rainwater at this location (Table 2). This suggests that inputs other than simple oxidation of ethanol are responsible for determining the concentration of the aldehyde in precipitation. The correlation was refined into winter and summer time periods and marine and continental back trajectories as well
Table 2 Correlation coefficients for various analytes in rainwater where correlations with p < 0.05 are considered significant and are represented in bold.
Ethanol Acetaldehyde
Acetaldehyde
NO 3
NSS
Hþ
DOC
0.178
0.564 0.592
0.563 0.598
0.0169 0.147
0.164 0.776
This study represents the first detailed analysis of the variability of ethanol and acetaldehyde concentrations in rainwater. Significant concentrations (nM) of ethanol and acetaldehyde were present in all samples indicating these analytes are ubiquitous components of precipitation at this location. There was a great deal of variability in the abundance of ethanol and acetaldehyde between rain events driven primarily by temporal and air mass back trajectory influences. The ratio of ethanol to acetaldehyde was at a minimum during periods of peak solar intensity underscoring the importance of alcohol oxidation by a photochemically generated oxidant in the gas or aqueous phase. The inverse correlation between the E:A ratio and H2O2 in rainwater suggests that ethanol will have a significant impact on the oxidizing and acid generating capacity of atmospheric waters. Results of this study are significant because they represent an important baseline for ethanol and acetaldehyde concentrations in North American rainwater. Comparison of future ethanol and acetaldehyde concentrations in the condensed phase to the data presented in this study will help delineate potential consequences of these labile OVOCs on the chemistry of the troposphere as the United States transitions to more ethanol blended fuels. Aqueous phase impacts of increasing ethanol concentrations will be particularly significant to the oxidizing capacity of atmospheric waters because of its reactivity with OH and HO2 radicals in solution. Increased rainwater concentrations could also have significant ramifications on receiving watersheds because of the biogeochemical lability of the alcohol.
References Avery, G.B., Farhana, B.K., Mead, R.N., Southwell, M., Willey, J.D., Kieber, R.J., 2013. Carbon isotopic characterization of hydrophobic dissolved organic carbon in rainwater. Atmos. Environ. 68, 230e234. Avery, G.B., Kieber, R.J., Witt, M., Willey, J.D., 2006. Rainwater monocarboxylic and dicarboxylic acid concentrations in southeastern North Carolina, USA as a function of air mass back trajectory. Atmos. Environ. 40, 1683e1693. Beale, R., Liss, P.S., Nightingale, P.D., 2010. First oceanic measurements of ethanol and propanol. Geophys. Res. Lett. 37. Blando, J.D., Turpin, B.J., 2000. Secondary organic aerosol formation in cloud and fog droplets: a literature evaluation of plausibility. Atmos. Environ. 34, 1623e1632. de Gouw, J.A., Gilman, J.B., Borbon, A., Warneke, C., Kuster, W.C., Goldan, P.D., Holloway, J.S., Peishl, J., Ryerson, T.B., Parrish, D.D., Gentner, D.R., Goldstein, A.H., Harley, R.A., 2012. Increasing atmospheric burden of ethanol in the United States. Geophys. Res. Lett. 39. Draxler, R.R., Rolph, G.D., 2003. HYSPLIT (Hybrid Single Particle Lagrangian Integrated Trajectory) Model. Real Time Environmental Display System. Ginnebaugh, D.L., Liang, J., Jacobson, M.Z., 2010. Examining the temperature dependence of ethanol (E85) versus gasoline emissions on air pollution with a largely-explicit chemical mechanism. Atmos. Environ. 44, 1192e1199. Kieber, R.J., Guy, A.L., Roebuck, J.A., Carroll, A.L., Mead, R.N., Jones, S.B., Giubbina, F.F., Campos, M.L.A.M., Willey, J.D., Avery, G.B., 2013. Determination of ambient ethanol concentrations in aqueous environmental matrices by two independent analyses. Anal. Chem.. Kieber, R.J., Parler, N.E., Skrabal, S.A., Willey, J.D., 2008. Speciation and photochemistry of mercury in rainwater. J. Atmos. Chem. 60, 153e168. Kieber, R.J., Rhines, M.F., Willey, J.D., Avery, G.B., 1999. Rainwater formaldehyde: concentration, deposition and photochemical formation. Atmos. Environ. 33, 3659e3667. Kieber, R.J., Whitehead, R.F., Willey, J.D., Reid, S., Seaton, P.J., 2006. Chromophoric dissolved organic matter (CDOM) in rainwater, southeastern North Carolina, USA. J. Atmos. Chem. 54, 21e41.
R.J. Kieber et al. / Atmospheric Environment 84 (2014) 172e177 Kieber, R.J., Willey, J.D., Zvalaren, S.D., 2002. Chromium speciation in rainwater: temporal variability and atmospheric deposition. Environ. Sci. Technol. 36, 5321e5327. Kieber, R.J., Zhou, X.,Z., Mopper, K., 1990. Formation of carbonyl compounds from UV-induced photodegradation of humic substances in natural waters: fate of riverine carbon in the sea. Limnol. Oceanogr. 35, 1503e1515. Kirstine, W.V., Galbally, I.E., 2012. The global atmospheric budget of ethanol revisited. Atmos. Chem. Phys. 12, 545e555. Lewis, A.C., Hopkins, J.R., Carpenter, L.J., Stanton, J., Read, K.A., Pilling, M.J., 2005. Sources and sinks of acetone, methanol, and acetaldehyde in North Atlantic marine air. Atmos. Chem. Phys. 5, 1963e1974. Millet, D.B., Apel, E., Henze, D.K., Hill, J., Marshall, J.D., Singh, H.B., Tessum, C.W., 2012. Natural and anthropogenic ethanol sources in North America and potential atmospheric impacts of ethanol fuel use. Environ. Sci. Technol. 46, 8484e8492. Millet, D.B., Guenther, A., Siegel, D.A., Nelson, N.B., Singh, H.B., de Gouw, J.A., Warneke, C., Williams, J., Eerdekens, G., Sinha, V., Karl, T., Flocke, F., Apel, E., Riemer, D.D., Palmer, P.I., Barkley, M., 2010. Global atmospheric budget of acetaldehyde: 3-D model analysis and constraints from in-situ and satellite observations. Atmos. Chem. Phys. 10, 3405e3425. Monod, A., Bonnefoy, N., Kalunzy, P., Denis, I., Foster, P., Carlier, P., 2003. Methods for sampling and analysis of tropospheric ethanol in gaseous and aqueous phases. Chemosphere 52, 1307e1319.
177
Mullaugh, K.M., Kieber, R.J., Willey, J.D., Avery, G.B., 2011. Long term temporal variability in hydrogen peroxide concentrations in Wilmington, North Carolina USA rainwater. Environ. Sci. Technol. 45, 9538e9542. Naik, V., Fiore, A.M., Horowitz, L.W., Singh, H.B., Wiedinmyer, C., Guenther, A., de Gouw, J.A., Millet, D.B., Goldan, P.D., Kuster, W.C., Goldstein, A., 2010. Observational constraints on the global atmospheric budget of ethanol. Atmos. Chem. Phys. 10, 5361e5370. Poulopoulos, S.G., Samaras, D.P., Philippopoulos, C., 2001. Regulated and unregulated emissions from an internal combustion engine operating on ethanolcontaining fuels. Atmos. Environ. 35, 4399e4406. Southwell, M.W., Smith, J.D., Avery, G.B., Kieber, R.J., Willey, J.D., 2010. Seasonal variability of formaldehyde production from photolysis of rainwater dissolved organic carbon. Atmos. Environ. 44, 3638e3643. Topol, L.E., Levon, M., Flanagan, J., Schwall, R.J., Jackson, A.E., 1985. Quality Assurance Management for Precipitation Systems. Environmental Protection Agency, Research Triangle Park, North Carolina. Warneck, P., 2006. A note on the temperature dependence of Henry’s law coefficients for methanol and ethanol. Atmos. Environ. 40, 7146e7151. Willey, J.D., Kieber, R.J., Eyman, M.S., Avery, G.B., 2000. Rainwater dissolved organic carbon: concentrations and global flux. Glob. Biogeochem. Cycles 14, 139e148.