Article pubs.acs.org/est
PAH Measurements in Air in the Athabasca Oil Sands Region Yu-Mei Hsu,*,†,‡ Tom Harner,§,‡ Henrik Li,¥,‡ and Phil Fellin¥,‡ †
Wood Buffalo Environmental Association, 100-330 Thickwood Boulevard, Fort McMurray, Alberta T9K 1Y1, Canada Environment Canada, Air Quality Processes Research Section, Toronto, Ontario M3H 5T4, Canada ¥ Airzone One Ltd., 222 Matheson Boulevard East, Mississauga, Ontario L4Z 1X1, Canada §
S Supporting Information *
ABSTRACT: Polycyclic aromatic hydrocarbon (PAH) measurements were conducted by Wood Buffalo Environmental Association (WBEA) at four community ambient Air quality Monitoring Stations (AMS) in the Athabasca Oil Sands Region (AOSR) in Northeastern Alberta, Canada. The 2012 and 2013 mean concentrations of a subset of the 22 PAH species were 9.5, 8.4, 8.8, and 32 ng m−3 at AMS 1 (Fort McKay), AMS 6 (residential Fort McMurray), AMS 7 (downtown Fort McMurray), and AMS 14 (Anzac), respectively. The average PAH concentrations in Fort McKay and Fort McMurray were in the range of rural and semirural areas, but peak values reflect an industrial emission influence. At these stations, PAHs were generally associated with NO, NO2, PM2.5, and SO2, indicating the emissions were from the combustion sources such as industrial stacks, vehicles, residential heating, and forest fires, whereas the PAH concentrations at AMS 14 (∼35 km south of Fort McMurray) were more characteristic of urban areas with a unique pattern: eight of the lower molecular weight PAHs exhibited strong seasonality with higher levels during the warmer months. Enthalpies calculated from Clausius−Clapeyron plots for these eight PAHs suggest that atmospheric emissions were dominated by temperature-dependent processes such as volatilization at warm temperatures. These findings point to the potential importance of localized water−air and/or surface−air transfer on observed PAH concentrations in air.
■
INTRODUCTION The Alberta oil sands are a mixture of sand, water, and bitumen that occur in the Athabasca, Peace River, and Cold Lake regions. The oil production rate in 2013 was 1.9 million barrels of oil per day with production expected to reach 4.8 million barrels of oil per day by 2030.1 The Athabasca Oil Sands Region (AOSR) in northeastern Alberta is the largest deposit, and approximately 20% of the resource is accessible by surface mining. The remaining deeper bitumen deposits are being recovered by in situ extraction methods,2 for example, Steam Assisted Gravity Drainage (SAGD). In recent years, there has been concern regarding the release of polycyclic aromatic compounds (PAC) from mining activities and potential effects on the Athabasca River and surrounding environment.3 Many studies have been carried out to characterize the PAC concentrations in the Athabasca River and its tributaries,4,5 including the assessment of fugitive emissions of polycyclic aromatic hydrocarbon (PAH) to air from tailings ponds.6−8 In ambient air, PAHs comprise a group of over 100 compounds that are produced as a result of incomplete combustion of which 17 PAHs have been identified as having adverse health effects.9 The Alberta Ambient Air Quality Objective for benzo(a)pyrene (BaP) is 0.3 ng m−3 (annual average). The World Health Organization indicates that the concentrations of BaP producing excess cancer risks of © XXXX American Chemical Society
1/10 000, 1/100 000 and 1/1 000 000 are 1.2, 0.12, and 0.012 ng m−3, respectively.10 In the AOSR, there are varied PAH emission sources including oil sands operations (e.g., fixed, mobile, and fugitive), natural (e.g., forest fires and vegetation), and community activities (e.g., transportation, BBQ, and wood burning). Wood Buffalo Environmental Association (WBEA) has utilized continuous air quality monitors and time-integrated measurements to monitor ambient air quality in AOSR since 1998 and currently operates 18 continuous ambient air quality monitoring stations. PAH data for the period 2012−2013 are now publically available at http://wbea.org/monitoringstations-and-data/integrated-data.11 As a result of the increased awareness and concerns for PAHs in the region, the objectives of this study are to characterize the PAH concentrations at the community ambient air quality monitoring stations in the AOSR and to identify the possible emission sources. Received: January 12, 2015 Revised: March 18, 2015 Accepted: April 6, 2015
A
DOI: 10.1021/acs.est.5b00178 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
■
METHODS Site Description. PAH measurements were conducted at four community stations: Air Monitoring Station 1 (AMS 1, 57.189N, 111.640W) in Fort McKay, AMS 6 (56.751N, 111.476W), AMS 7 (56.732N, 111.390W) in Fort McMurray, and AMS 14 (56.449N, 111.037W) in Anzac (Figure 1). AMS
Fort McMurray, and Anzac were 562, 61 374, and 585 persons, respectively (www.statcan.gc.ca). Sampling and Analysis Method. U.S. Environment Protection Agency Compendium Method TO-13A12 has been applied to measure ambient PAH concentrations. The samples were collected by a Polyurethane Foam (PUF) Sampler and the sampling train consisted of glass fiber filter followed by a PUF plug in a glass cartridge. Sampling was carried out for 24 h every 6 days based on the Environment Canada National Air Pollution Surveillance network monitoring schedule from January 2012 to December 2013. A gas chromatograph with mass selective detector was employed to analyze 23 species, naphthalene (NAP), acenaphthylene (ACY), acenaphthene (ACE), fluorene (FLU), phenanthrene (PHE), anthracene (ANT), acridine (ACR), fluoranthene (FLT), pyrene (PYR), benzo(c)phenanthrene (BcP), benz(a)anthracene (BaA), chrysene (CRY), 7,12-dimethylbenz(a)anthracene (DBA), benzo(b,j)fluoranthene (BbjF), benz(k)fluoranthene (BkF), benzo(a)pyrene (BaP), 3-methylcholanthrene (MCA), indeno(1,2,3-cd)pyrene (IND), dibenz(a,h)anthracene (dBahA), benzo(ghi)perylene (BghiP), dibenzo(a,l)pyrene (dBalP), dibenzo(a,i)pyrene (dBaiP), and dibenzo(a,h)pyrene (dBahP). The sampling medium, the PUF plug, is well-known with low collection efficiencies for NAP;13 hence, the NAP concentration is not reported in this study. For ACY and ACE, the collection efficiencies were verified by using two PUF plugs (size of samples = 88), and the average collection efficiencies (±standard deviation) were 78% (±20%) for ACY and 75% (±16%) for ACE. In this study, the results (measurement data) were not corrected for efficiency. Hourly concentrations of nitrogen monoxide (NO) with nitrogen dioxide (NO2), sulfur dioxide (SO2), total hydrogen carbon (THC), PM2.5 (particulate matter with aerodynamic diameter smaller than 2.5 μm), and ambient air temperature were monitored by an NO/NO2 analyzer, SO2 analyzer, THC analyzer, Federal Equivalent Method PM2.5 monitor, and a temperature probe, respectively. PM10 (particulate matter with aerodynamic diameter smaller than 10 μm) samples were collected by a Federal Reference Method PM10 sampler with sulfate (SO42−), nitrate (NO3−) and ammonium (NH4+) analyzed by an ion chromatography system and metal species, includes aluminum (Al), iron (Fe), vanadium(V), analyzed by inductively coupled plasma-mass spectrometry. Quality Control. For PAH sampling, one field blank and four samples were collected on each sampling day. For PAH analysis, solvent blank, laboratory blank, and replicate analysis for every four samples was performed. Phenanthrene-d10 and chrysene-d12 were used as extraction internal standards. Fluoranthene-d10 and pyrene-d10 were used as instrument performance internal standards. Instrument detection limit (IDL), extraction recovery efficiency, method detection limits (MDL), PAH concentrations of field blank samples and relative percent difference of replicate analyses are listed in Supporting Information (SI) Table S1 to S3. All data were field-blank subtracted. Monthly multipoint calibration and daily zero-span (two-point calibration with zero gas for the baseline and span gas for the upper limit of a gas concentration measurement) were carried out for continuous ambient air quality monitors, including monitors for NO/NO2, SO2, and THC following Alberta’s Air Monitoring Directive and the standard operation procedures are publicly available at www.wbea.org.
Figure 1. Satellite image of a portion of the AOSR showing the locations of the four continuous ambient air quality monitoring stations (Google map, October 15, 2014).
1 is located in the First Nation and Metis Community of Fort McKay, which is approximately within 15 km of four mining/ upgrading operations. Both AMS 6 (residential area) and AMS 7 (Athabasca river valley in downtown) are located in Fort McMurray, 35 km to the south of these two oil sands operations. The distance between AMS 6 and AMS 7 is approximately 6 km. The station at AMS 14 was established to monitor air quality for the local community of Anzac, located ∼35 km to the southeast of Fort McMurray. In the region around AMS 14 and within ∼6 km of the sampling site, the SAGD process with an associated bitumen upgrader is used to recover the bitumen. The 2011 populations of Fort McKay, B
DOI: 10.1021/acs.est.5b00178 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
Figure 2. 2012 and 2013 monthly VPAH, SVPAH, and PPAH concentrations at AMS 1, AMS 6, AMS 7, and AMS 14 (VPAH = Volatile PAHs, including ACY, ACE, FLU, PHE, and ANT; SVPAH = Semivolatile PAHs, including ACR, FLT, PYR, BcP, BaA, and CRY; PPAH = Particulate PAH, including DBA, BbjF, BkF, BaP, MCA, IND, dBahA, BghiA, dBalP, dBaiP, and dBahP).
■
Temporal Variation. The twenty-two PAH species were sorted into three groups, as shown in SI Table S4, according to their vapor pressure for the purpose of temporal variation analysis. The three groups are (1) volatile PAHs (VPAH): ACY, ACE, FLU, PHE and ANT; (2) semivolatile PAHs (SVPAH): ACR, FLT, PYR, BcP, BaA, and CRY; and (3) particulate PAHs (PPAH): DBA, BbjF, BkF, BaP, MCA, IND, dBahA, BghiA, dBalP, dBaiP, and dBahP. Monthly concentrations of VPAH, SVPAH, and PPAH are displayed in Figure 2. At AMS 1, the VPAH concentrations during winter months were slightly higher than for summer months, except in June 2012, due to the influence from a forest fire event. The VPAH concentrations at AMS 1, AMS 6 and 7 were similar and positively correlated with each other (r = 0.54 for AMS 1 and AMS 6; 0.48 for AMS 1 and AMS 7; 0.50 for AMS 6 and AMS 7, P < 0.01). The VPAH concentrations at AMS 14 were significantly higher compared to the other three sites and showed a distinct and strong seasonal pattern that was not observed at the other sites. The AMS 14 correlation coefficients were 0.79 (P < 0.01) for the VPAH concentration and ambient temperature (Table 1) for two sampling years and −0.43 (P < 0.01) (SI Table S7) for the VPAH concentration and wind speed for summer months, from April to October. SVPAH concentrations (Figure 2e−h) and patterns were also similar among the three sites AMS 1, AMS 6, and AMS 7 and were correlated well with one another (r = 0.50 for AMS 1 and AMS 6; 0.41 for AMS 1 and AMS 7; 0.51 for AMS 6 and AMS 7, P < 0.01). SVPAH concentrations at AMS 1 were more variable and slightly higher than at AMS 6 and AMS 7. At AMS 14, the SVPAH concentrations showed the strong seasonal pattern that was observed for the VPAH. The correlation coefficient for the SVPAH concentration and ambient temperature was 0.72 (P < 0.01, Table 1). The unique seasonal pattern for VPAH and SVPAH at AMS14 indicates a distinct and seasonally dependent emission source (particularly for the
RESULTS AND DISCUSSION Spatial Variation. The average concentrations of the PAHs considered in this analysis in 2012 and 2013 at four stations are listed in SI Table S4. The PAH concentrations were dominated by low-molecular-weight PAHs including, ACY, ACE, FLU, PHE, ANT, ACR, FLT, and PYR at all sites. The 2012−2013 average concentrations (±standard deviation) of the total of 22 PAHs were 9.5 ± 8.8, 8.4 ± 7.3, 8.8 ± 5.7, and 32 ± 47 ng m−3 at AMS 1, AMS 6, AMS 7 and AMS 14, respectively. The PAH concentrations were similar at AMS 1, AMS 6, and AMS 7 (P > 0.05, SI Table S5). However, the concentrations of PAHs including, ACE, FLU, PHE, ANT, ACR, and FLT at AMS 14 were considerably higher compared to the other three sites (P < 0.05, SI Table S5); however, the concentrations of the higher molecular weight PAHs, BcP, BaA, CRY and BaP were lower at AMS14 compared to the other three sites (P < 0.05, SI Table S5). The mean concentrations (±standard deviation) of ACY, FLU, and PHE (major three species) were 1.1 ± 2.2, 1.3 ± 1.3, and 3.5 ± 3.1 ng m−3 at AMS 1; 1.2 ± 2.4, 1.2 ± 0.93, and 2.9 ± 2.2 ng m−3 at AMS 6; 1.3 ± 1.6, 1.5 ± 1.1, and 2.9 ± 1.8 ng m−3 at AMS 7 (SI Table S4). For AMS 14, the concentrations (mean ± standard deviation) of four major species were 14 ± 22 ng m−3 for PHE, 7.7 ± 12 ng m−3 for FLU, 5.9 ± 11 ng m−3 for ACE, and 1.1 ± 1.6 ng m−3 for FLT. The 2012 and 2013 annual mean concentrations of BaP were highest in Fort McKay, that is, 0.063 and 0.066 ng m−3 at AMS 1; 0.041 and 0.046 ng m−3 at AMS 6; 0.039 and 0.041 ng m−3 at AMS 7; 0.021 and 0.028 ng m−3 at AMS 14. In general, the PAH concentrations at AMS 1, AMS 6 and AMS 7 were lower than for urban areas (Chicago and Cleveland in the U.S.; Hamilton and Toronto in CA; London and Manchester in the U.K., SI Table S6) and within the range of rural and semirural areas, but the concentrations at AMS 14 were similar to or higher than urban areas.14−20 Elevated PAHs at AMS 14 were also reported from monitoring using passive air samplers.21,22 C
DOI: 10.1021/acs.est.5b00178 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
14 7 6
−0.17 0.06 0.08 0.36 −0.06 0.50 119
1
−0.10 0.26 0.04 0.66 −0.14 0.12 119
14
0.24
