Environ Monit Assess (2015) 187:474 DOI 10.1007/s10661-015-4669-1
Distribution of polycyclic aromatic hydrocarbons in sediments of Akaki River, Lake Awassa, and Lake Ziway, Ethiopia Kebede Nigussie Mekonnen & Bhagwan Singh Chandravanshi & Mesfin Redi-Abshiro & Abayneh Ataro Ambushe & Robert Ian McCrindle & Stanley Moyo
Received: 9 March 2015 / Accepted: 3 June 2015 # Springer International Publishing Switzerland 2015
Abstract The quantification of 14 polycyclic aromatic hydrocarbons (PAHs) was carried out in sediment samples collected from Akaki River, Lake Awassa, and Lake Ziway, Ethiopia. The concentration of PAHs in the samples was determined using gas chromatographymass spectrometry (GC-MS) in selected ion monitoring (SIM) mode, after microwave-assisted extraction (MAE), using acetone/n-hexane (1:1, v/v) mixture. The
Electronic supplementary material The online version of this article (doi:10.1007/s10661-015-4669-1) contains supplementary material, which is available to authorized users. K. N. Mekonnen : B. S. Chandravanshi : M. Redi-Abshiro (*) Department of Chemistry, Addis Ababa University, P. O. Box 1176, Addis Ababa, Ethiopia e-mail: [email protected]
accuracy of the method was determined by extracting and analyzing New York/New Jersey waterway sediment standard reference material (SRM 1944). The measured concentrations of PAHs in SRM 1944 agreed well with the certified values. In samples from Akaki River, Lake Awassa, and Lake Ziway, the total content of PAHs determined ranged from 0 to 3070 ng/g (average 534 ng/g), 24.9 to 413 ng/g (average 169 ng/g), and 15.0 to 305 ng/g (average 175 ng/g), respectively. Source ratios indicated that the PAHs were mainly from petrogenic origin. Sediments from all sampling sites indicated negligible levels of toxicity with no risk of adverse biological effects. Keywords Sediment . Polycyclic aromatic hydrocarbons . Microwave-assisted extraction . Gas chromatography-mass spectrometry
M. Redi-Abshiro e-mail: [email protected]
Introduction K. N. Mekonnen Department of Chemistry, Mekelle University, P. O. Box 231, Mekelle, Ethiopia K. N. Mekonnen : R. I. McCrindle Department of Chemistry, Tshwane University of Technology, P. O. Box 56208, Arcadia 0007, South Africa A. A. Ambushe Department of Chemistry, University of Limpopo, Private Bag x1106, Sovenga 0727, South Africa S. Moyo School of Chemistry, University of the Witwatersrand, Private Bag 3, P. O. Box Wits 2050, Johannesburg, South Africa
Polycyclic aromatic hydrocarbons (PAHs) are organic compounds containing two or more fused aromatic rings (Qiao et al. 2006; Yim et al. 2007; Nikolaou et al. 2009; Hahladakis et al. 2013). The number and position of the rings, as well as the number, position, and nature of the functional groups, affect the physical and chemical properties, environmental behavior, and interactions with biota and humans (Nikolaou et al. 2009). They originate from anthropogenic (fossil fuel combustion, municipal and industrial effluents, oil spills, agricultural and urban runoff, vehicle exhausts, asphalt production,
474
Page 2 of 13
and waste incineration) and natural sources (all incomplete natural combustion at high temperature and pyrolytic processes involving fossil fuels, forest fires, and volcanic eruptions) (Martinez et al. 2004; Culotta et al. 2006; Li et al. 2006; Qiao et al. 2006; Liang et al. 2007; Yim et al. 2007; Orecchio and Papuzza 2009; Hahladakis et al. 2013). The PAHs are ubiquitous environmental contaminants, present in complex mixtures, which makes human and environmental exposure unavoidable (Qiao et al. 2006; Duke and Albert 2007; Orecchio et al. 2009). Since they are widespread in all environmental compartments, the level of PAHs in environmental samples has resulted in interest from analytical chemists (Shu et al. 2003). Different PAHs may induce adverse effects, such as immunotoxicity, genotoxicity, carcinogenicity, and reproductive toxicity (Sverdrup et al. 2002; Qiao et al. 2006). The US Environmental Protection Agency (USEPA) classified PAHs as priority pollutants because of their carcinogenic and mutagenic properties. In particular, benzo[a]anthracene, chrysene, benzo[b]fluoranthene, benzo[a]pyrene, and benzo[ghi]perylene are considered human carcinogens (Duke and Albert 2007; Liang et al. 2007; Wang et al. 2007; Nikolaou et al. 2009). They may also synergistically increase the carcinogenicity of other organic compounds (Orecchio and Papuzza 2009; Orecchio et al. 2009). Sediments are usually the ultimate sinks of PAHs and other contaminants discharged into the environment. They reflect the input from point and non-point sources of contamination (Yamada et al. 2009; Hahladakis et al. 2013; Yohannes et al. 2013a, b). Properties, such as high toxicity, high stability, high lipophilicity, electrochemical stability, and adsorption to sediments, make PAHs a potentially dangerous group of chemicals. The sediment becomes a long-term repository and a steady indicator of environmental pollution (Martinez et al. 2004; Villar et al. 2004; Duke and Albert 2007; Kumar et al. 2008; Nikolaou et al. 2009; Hahladakis et al. 2013). Once PAH-enriched particles accumulate in sediments, they may undergo a number of changes caused by chemical, biological, and physical activities. As a result, the bound PAHs can be remobilized from the sediment into the water phase and tend to bioaccumulate in aquatic organisms (Kumar et al. 2008). The PAHs are commonly determined by capillary gas chromatography, with flame ionization detection (GCFID), mass spectrometric detection (gas chromatography-mass spectrometry (GC-MS)), or by
Environ Monit Assess (2015) 187:474
high-performance liquid chromatography (HPLC) with ultraviolet (UV) or fluorescence detection (Martinez et al. 2004; Denis et al. 2012). The use of GC-MS for determination of PAHs is based on a favorable combination of greater selectivity, resolution, and sensitivity (Alvarez-Aviles et al. 2007; Yamada et al. 2009; Denis et al. 2012). However, due to the complex nature of the sediments, prior to introduction into the analytical system, a sample preparation (extraction and cleanup) process is needed (Shu et al. 2000; Alvarez-Aviles et al. 2007; Yamada et al. 2009). Sample extraction is a critical step in the determination of PAH because these compounds are strongly sorbed to the matrix; consequently, their extraction is time-consuming and, in many cases, causes quantification errors (Yamada et al. 2009). The extraction of PAHs from sediments can be achieved either with traditional or advanced methods. The traditional techniques require long extraction times, consume large amounts of solvent, and can degrade thermally labile compounds (Dean and Xiong 2000; Shu et al. 2003; Villar et al. 2004; Alvarez-Aviles et al. 2007; Wang et al. 2007; Chan et al. 2011). Currently, advanced extraction techniques, such as supercritical fluid extraction (SFE), accelerated solvent extraction (ASE), and microwave-assisted extraction (MAE), have been established in order to improve the drawbacks of traditional extraction techniques (Dean and Xiong 2000; Shu et al. 2000, 2003; Villar et al. 2004; Alvarez-Aviles et al. 2007; Wang et al. 2007; Nikolaou et al. 2009; Chan et al. 2011). The MAE is gaining an important role for sample preparation because it offers greatly reduced usage of organic solvents, reduced extraction times, and reduced size of extraction apparatus and permits simultaneous extraction of multiple samples (Dean and Xiong 2000; Villar et al. 2004; Wang et al. 2007; Chan et al. 2011). Two major rivers flow through Addis Ababa, Ethiopia, namely the Tinishu Akaki and Tiliku Akaki. These rivers carry domestic waste, rain runoff water, and untreated industrial discharges of various types, hence making them known for their offensive odor (Xu and Usher 2006; Prabu 2009). Lake Awassa and Ziway are found in the Ethiopian Rift Valley. The threats to the lakes are factory effluents, solid and liquid wastes generated by different sectors, small- and large-scale farming, and also sewer lines (Awulachew et al. 2007; Yohannes et al. 2013a, b). With regard to the status of potentially toxic elements, we recently reported that the sediments collected from the Akaki River was polluted
Environ Monit Assess (2015) 187:474
moderate or to a heavy degree by Cr, Zn, and Pb, notpolluted to moderate rate with Ni, and non-polluted by Cu (Mekonnen et al. 2014). Similarly, we also reported that Lake Awassa and Lake Ziway were not to moderately polluted with Cr and Ni but highly polluted with Zn (Mekonnen et al. 2015). However, no investigations have been carried out on the distribution and sources of PAHs in these important ecosystems. Yohannes et al. (2013a, b) investigated the levels of organochlorine pesticides (OCPs) and heavy metals in fish and sediment collected from Lake Awassa but not PAHs. Prasse et al. (2012) assessed contamination and sources of metals, polychlorinated biphenyles (PCBs), and PAHs in urban soils from Addis Ababa, Ethiopia, but not water systems. The objectives of this investigation were to assess the concentration, distribution, and sources of PAHs in sediments of the Akaki River, Lake Awassa, and Lake Ziway in Ethiopia using GC-MS after MAE and cleanup steps. In addition, ecological risks of these contaminants in surface sediments were also assessed by comparison with sediment quality guidelines (SQGs).
Experimental Chemicals and reagents For the extractions, 99.8 % acetone and 95 % n-hexane (HPLC grade, LAB-SCAN Analytical Sciences, Poland) were used. Copper metal powder was obtained from UNILAB (Philip Harris, UK) and activated using 99.9 % methanol (HPLC grade, LAB-SCAN Analytical Sciences, Poland), dichloromethane, and diethyl ether (AAR grade, SMM Instruments, South Africa). Anhydrous sodium sulfate (analyzed reagent >99.0 % UNIVAR) was obtained from Merck Chemicals (Pty) Ltd, Germany. Argon (99.998 % purity) and helium (99.999 %, purity) were supplied by Afrox Boc gases (Afrox, South Africa). The PAH kit containing 16 priority pollutants (Supelco Inc., Bellefonte, PA, USA) was used for preparation of series of standards for external calibration. A Standard Reference Material of New York-New Jersey Waterway Sediment (SRM 1944) (National Institute of Standards and Technology (NIST), Gaithersburg, MD, USA) was used to evaluate the accuracy and precision of the method. All the glassware and polyethylene bottles, prior to use, were washed with detergent, rinsed with doubly deionized water (18.2 M
Page 3 of 13 474
Ω/cm, Millipore purification system, Millipore, France), soaked in 10 % v/v HNO3 (65 % Suprapur®, Merck, Germany) for 24 h, and dried in an oven. Sample collection and preparation Three sampling areas, namely Addis Ababa, Awassa, and Ziway, were selected. From Addis Ababa, 15 sampling sites directly and/or indirectly related to Akaki River (AA Bulbula, AA Entoto, AA Gefersa, AA Kebena, AA Kera, AA Kidanemihret before TAR, AA TAR Alert, AA TAR before Kidanemihret, AA TAR before Melkaqurani, AA TAR Kaliti, AA Melkaqurani before TAR, AA TAR Kolfe, AA TAR with Kidanemihret, AA TlAR at KK Textile Factory, and AA TlAR with TAR at Aba Samuel) were selected for collecting sediment samples. Similarly, 11 sites from Awassa (Amoragedel, Asamaber, Buko, Fidelserawit, Kereisa, Mezinagna, Muate, Samijersa, Shewaber, Tikurwuha, and Wondotika) and ten sites from Ziway (Bochesa, Debresina, Gabriel, Gelila, Korekonch, Mezinagna, Shalo, Sedeicha, Wameicha, and Wofdeset) were selected (the maps of the sampling areas with the sampling points are attached as supplementary documents). Sediment samples were taken at a depth of 20 cm by a stainless steel grab sampler using 500-mL bottles which were kept in a cool container. Samples were air-dried, homogenized using an agate mortar and pestle and passed through a sieve (100 mesh), and stored until analysis. Microwave-assisted extraction A CEM MARS Xpress Microwave Accelerated Reaction System (CEM Corporation, Matthews, NC, USA) was used for extraction of PAHs. For extraction, EPA Method 3546 (microwave extraction) was used with some modifications and optimized for PAH extraction as follows. About 2-g sediment samples were weighed into extraction vessels, and extraction was done with the optimized method using 30 mL of mixture of n-hexane and acetone (1:1, v/v) for 30 min at a 110 °C at 200 W. After the vessels were allowed to cool to room temperature, the supernatant was filtered through a precleaned Pasteur’s pipette glass fitted with solvent-rinsed glass wool. Then, the solvent was decanted and subjected to a cleanup procedure (EPA Method 3660, sulfur cleanup). To the extract, 2 g of precleaned anhydrous Na2SO4 was added, mixed thoroughly, left overnight, and filtered. To
474
Environ Monit Assess (2015) 187:474
Page 4 of 13
the filtrate, 0.2 g of activated copper was added for desulfurization (Ledgard 2007), filtered with a premium syringe filter (diameter 0.22 μm, Agilent), and evaporated to nearly dryness using a Genevac EZ-2 Series Personal Evaporator (Genevac Inc, UK). To the concentrated extract, 1 mL of dichloromethane was added and transferred to 1.5-mL brown vials for GC-MS analysis.
The gas chromatography-mass spectrometry instrumentation and condition The PAHs in the purified extract were determined using an Agilent 7890A series gas chromatograph interfaced with an inert Agilent 5975C mass selective detector (MSD) with triple-axis detector (Agilent Technologies, USA) operated in electronic impact mode at 70 eV. Separation of the compounds was achieved using a fused silica capillary column (DB-5 MS, 30 m×0.25mm i.d.×0.25-μm film thickness; J&W Scientific, Folsom, CA, USA). The instrumental parameters are shown in Table 1.
Identification and quantification The identity of the PAHs in the sediment samples was confirmed by retention time and abundance of confirmation ions in the authentic PAHs standards. The USEPA 14 priority PAHs were quantified based on a minimum of seven-point calibration curve for individual compounds. Automated library searching was performed using the NIST Mass Spectral Database (Gaithersburg, MD, USA). The concentrations of PAHs are expressed on a dry mass basis.
Table 1 Operating conditions for GC-MS analysis Parameters
Description
Mobile phase
He
Column flow rate
0.8 mL/min
Oven temperature
Start at 60 °C and hold for 2 min, ramp to 160 °C at 15 °C/min and hold for 9 min, ramp to 260 °C at 5 °C/min and hold for 3 min, and then ramp to 300 °C at 5 °C/min and hold for 1.67 min with a total oven temperature run time of 50.34 min
MSD transfer line heater temperature Ion source temperature Quadrupole mass analyzer temperature Acquisition mode
310 °C 230 °C 150 °C
Identification in scan mode and quantification in selected ion monitoring (SIM) mode
Ions selected for each naphthalene (128, 127, 129, 102), PAH (m/z) acenaphthylene (152, 151, 76, 150), fluorene (166, 165, 167, 182), phenanthrene (178, 176, 179, 76), anthracene (178, 176, 179, 89), fluoranthene (202, 200, 203, 187), pyrene (202, 203, 101, 200), chrysene (228, 226, 229, 113), benzo[a]anthracene (228, 226, 229, 114), benzo[k]fluoranthene (252, 250, 253, 126), benzo[b]fluorantene (252, 253, 250, 126), benzo[a]pyrene (252, 253, 250, 126), indeno[1,2,3cd]pyrene (276, 277, 274, 136), and benzo[ghi]perylene (276, 137, 138, 274) The italicized QIons were used for quantification
Results and discussion Quality assurance
Optimization of microwave-assisted extraction
Analysis of the set of samples was accompanied by a procedural blank and a duplicate sample, which was analyzed in a similar fashion to the samples. The SRM 1944 was analyzed for quality assurance checks for the modified extraction EPA Method 3546 and cleanup procedure. Instrumental calibrations were checked by the injection of a continuing calibration solution (0.5 μg/ mL concentration). The GC-MS calibration was verified by running ten samples.
Microwave-assisted extraction In order to perform a simple, fast, and efficient extraction of PAHs, the amount of solvents, microwave power, and extraction time were studied. The literature indicated that the highest PAH recoveries were obtained using hexane/acetone (1:1, v/v) (Shu et al. 2000; Villar et al. 2004; Li et al. 2006; Svoboda et al. 2007; Wang et al. 2007). This solvent was therefore used for extraction.
Environ Monit Assess (2015) 187:474
Page 5 of 13 474
Sulfur is considered to be an interfering element in PAH analysis in sediment since it is soluble in organic solvents and co-extracted with target organic compounds. Removal of sulfur is possible using different reagents: potassium hydroxide in ethanol (Ricklund et al. 2008), tetrabutylammonium (TBA) hydrogen sulfate saturated with sodium sulfite (Svoboda et al. 2007), and activated copper (Ledgard 2007; Liang et al. 2007). Activated copper was therefore used for removal of sulfur from sediment samples. Method validation The relative standard deviation of duplicate samples was less than 10 %. The efficiency of MAE and the cleanup of sediment samples and sample blanks were checked using SRM 1944. The results indicated satisfactory recovery, with the average recovery ranging from 89.0 to 98.2 % (Table 2). The detection and quantification limits were set at three and ten times the background noise obtained for blank samples, respectively. The detection limit ranged from 1.28 ng/g (phenanthrene) to 3.92 ng/g (naphthalene), while the quantification limit ranged from Table 2 Certified and measured concentrations for selected PAHs in SRM 1944 for method validation PAHs
Certified concentration values (mg/kg dry mass)
Measured concentration (mg/kg dry mass, n=3)
Naphthalene
1.65±0.31
1.53±0.17
Acenaphthylene
NA
–
Fluorenea
0.85±0.03
0.77±0.06
Phenanthrene
5.27±0.22
5.11±0.16
Anthracene
1.77±0.33
1.63±0.19
Fluoranthene
8.92±0.32
8.53±0.27
Pyrene
9.70±0.42
8.88±0.41
Benz[a]anthracene
4.72±0.11
4.20±0.29
Chrysene
4.86±0.10
4.63±0.13
Benzo[b]fluoranthene
3.87±0.42
3.73±0.57
Benzo[k]fluoranthene
2.30±0.20
2.23±0.18
Benzo[a]pyrene
4.30±0.13
4.22±0.17
Indeno[1,2,3-cd]pyrene
2.78±0.10
2.73±0.15
Benzo[ghi]perylene
2.84±0.10
2.63±0.11
NA data not available a
This concentration is provided as a reference value
4.27 ng/g (phenanthrene) to 13.1 ng/g (naphthalene) (Table 3). The repeatability of at least two parameters in the gas chromatographic method: the retention time (confirming the identity of the analyte of interest) and the peak area or height (quantifying the analyte of interest) were achieved satisfactorily with the chromatographic method under the study conditions. Analytical determination The complete resolution of the 14 USEPA priority PAHs was obtained with a careful choice of suitable chromatography conditions, including the stationary phase, by carrying out a fast GC analysis on a DB-5 MS fused silica capillary column after MAE and cleanup. Quantification was performed using at least a seven-point calibration curve for each individual PAH. The correlation coefficient (R 2 ) varied from 0.955 to 0.998 (Table 3). However, for the two PAHs (acenaphthene and dibenzo(a,h)anthracene), the calibration curves were not sufficiently satisfactory to continue the work on the compound, and hence, only the remaining 14 PAHs were studied. The residual characteristics of PAH in studied area The total concentrations of PAHs (dry mass basis) are presented in Table 4. The highest total concentration of PAHs was found in the Akaki River (3070 ng/g) samples taken from at the TAR Alert area, while the lowest concentration ( phenanthrene (20 %) > fluoranthene (19 %) > pyrene (14 %) > anthracene (10 %) > fluorene (5 %) > benzo[ghi]perylene (2 %) ≈ acenaphthylene (2 %) > benzo[b]fluorantene (1 %) ≈ indeno[1,2,3-cd]pyrene (1 %) ≈ benzo[a]pyrene (1 %) ≈ chrysene (1 %) ≈ benzo[k]fluoranthene (1 %) > benzo[a]anthracene (0.3 %); the distribution varied substantially among the different locations. Therefore, of the 14 PAHs, naphthalene, phenanthrene, and fluoranthene were predominant species (accounting for 63 %) while benzo[a]pyrene,
474
Environ Monit Assess (2015) 187:474
Page 6 of 13
Table 3 List of PAHs for analysis, correlation coefficient, detection, and quantification limits for SIM GC-MS mode PAHs
Retention time, min
R2
Detection limit (ng/g, dm)
Quantification limit (ng/g, dm)
Chemical formula
No. of ring
Naphthalene
C10H8
2
7.869
0.994
3.92
13.1
Acenaphthalene
C12H8
3
10.998
0.995
1.46
4.87
Fluorene
C13H10
3
13.564
0.991
1.53
5.10
Phenanthrene
C14H10
3
20.106
0.983
1.28
4.27
Anthracene
C14H10
3
20.485
0.998
2.22
7.40
Fluoranthene
C16H10
4
27.515
0.993
2.77
9.23
Pyrene
C16H10
4
28.908
0.991
1.78
5.93
Benza[a]anthracene
C18H12
4
35.187
0.987
3.38
11.3
Chrysene
C18H12
4
35.007
0.978
3.38
11.3
Benzo[b]fluoranthene
C20H12
5
40.503
0.981
2.69
8.97
Benzo[k]fluoranthene
C20H12
5
40.656
0.976
2.69
8.97
Benzo[a]pyrene
C20H12
5
42.232
0.987
2.55
8.50
Indeno[1,2.3-cd]pyrene
C22H12
6
47.549
0.955
2.33
7.77
Benzo[ghi]perylene
C22H12
6
48.251
0.990
1.99
6.63
dm dry mass
chrysene, benzo[k]fluoranthene, and benzo[a]anthracene were found the least (accounting
Distribution of polycyclic aromatic hydrocarbons in sediments of Akaki River, Lake Awassa, and Lake Ziway, Ethiopia Kebede Nigussie Mekonnen & Bhagwan Singh Chandravanshi & Mesfin Redi-Abshiro & Abayneh Ataro Ambushe & Robert Ian McCrindle & Stanley Moyo
Received: 9 March 2015 / Accepted: 3 June 2015 # Springer International Publishing Switzerland 2015
Abstract The quantification of 14 polycyclic aromatic hydrocarbons (PAHs) was carried out in sediment samples collected from Akaki River, Lake Awassa, and Lake Ziway, Ethiopia. The concentration of PAHs in the samples was determined using gas chromatographymass spectrometry (GC-MS) in selected ion monitoring (SIM) mode, after microwave-assisted extraction (MAE), using acetone/n-hexane (1:1, v/v) mixture. The
Electronic supplementary material The online version of this article (doi:10.1007/s10661-015-4669-1) contains supplementary material, which is available to authorized users. K. N. Mekonnen : B. S. Chandravanshi : M. Redi-Abshiro (*) Department of Chemistry, Addis Ababa University, P. O. Box 1176, Addis Ababa, Ethiopia e-mail: [email protected]
accuracy of the method was determined by extracting and analyzing New York/New Jersey waterway sediment standard reference material (SRM 1944). The measured concentrations of PAHs in SRM 1944 agreed well with the certified values. In samples from Akaki River, Lake Awassa, and Lake Ziway, the total content of PAHs determined ranged from 0 to 3070 ng/g (average 534 ng/g), 24.9 to 413 ng/g (average 169 ng/g), and 15.0 to 305 ng/g (average 175 ng/g), respectively. Source ratios indicated that the PAHs were mainly from petrogenic origin. Sediments from all sampling sites indicated negligible levels of toxicity with no risk of adverse biological effects. Keywords Sediment . Polycyclic aromatic hydrocarbons . Microwave-assisted extraction . Gas chromatography-mass spectrometry
M. Redi-Abshiro e-mail: [email protected]
Introduction K. N. Mekonnen Department of Chemistry, Mekelle University, P. O. Box 231, Mekelle, Ethiopia K. N. Mekonnen : R. I. McCrindle Department of Chemistry, Tshwane University of Technology, P. O. Box 56208, Arcadia 0007, South Africa A. A. Ambushe Department of Chemistry, University of Limpopo, Private Bag x1106, Sovenga 0727, South Africa S. Moyo School of Chemistry, University of the Witwatersrand, Private Bag 3, P. O. Box Wits 2050, Johannesburg, South Africa
Polycyclic aromatic hydrocarbons (PAHs) are organic compounds containing two or more fused aromatic rings (Qiao et al. 2006; Yim et al. 2007; Nikolaou et al. 2009; Hahladakis et al. 2013). The number and position of the rings, as well as the number, position, and nature of the functional groups, affect the physical and chemical properties, environmental behavior, and interactions with biota and humans (Nikolaou et al. 2009). They originate from anthropogenic (fossil fuel combustion, municipal and industrial effluents, oil spills, agricultural and urban runoff, vehicle exhausts, asphalt production,
474
Page 2 of 13
and waste incineration) and natural sources (all incomplete natural combustion at high temperature and pyrolytic processes involving fossil fuels, forest fires, and volcanic eruptions) (Martinez et al. 2004; Culotta et al. 2006; Li et al. 2006; Qiao et al. 2006; Liang et al. 2007; Yim et al. 2007; Orecchio and Papuzza 2009; Hahladakis et al. 2013). The PAHs are ubiquitous environmental contaminants, present in complex mixtures, which makes human and environmental exposure unavoidable (Qiao et al. 2006; Duke and Albert 2007; Orecchio et al. 2009). Since they are widespread in all environmental compartments, the level of PAHs in environmental samples has resulted in interest from analytical chemists (Shu et al. 2003). Different PAHs may induce adverse effects, such as immunotoxicity, genotoxicity, carcinogenicity, and reproductive toxicity (Sverdrup et al. 2002; Qiao et al. 2006). The US Environmental Protection Agency (USEPA) classified PAHs as priority pollutants because of their carcinogenic and mutagenic properties. In particular, benzo[a]anthracene, chrysene, benzo[b]fluoranthene, benzo[a]pyrene, and benzo[ghi]perylene are considered human carcinogens (Duke and Albert 2007; Liang et al. 2007; Wang et al. 2007; Nikolaou et al. 2009). They may also synergistically increase the carcinogenicity of other organic compounds (Orecchio and Papuzza 2009; Orecchio et al. 2009). Sediments are usually the ultimate sinks of PAHs and other contaminants discharged into the environment. They reflect the input from point and non-point sources of contamination (Yamada et al. 2009; Hahladakis et al. 2013; Yohannes et al. 2013a, b). Properties, such as high toxicity, high stability, high lipophilicity, electrochemical stability, and adsorption to sediments, make PAHs a potentially dangerous group of chemicals. The sediment becomes a long-term repository and a steady indicator of environmental pollution (Martinez et al. 2004; Villar et al. 2004; Duke and Albert 2007; Kumar et al. 2008; Nikolaou et al. 2009; Hahladakis et al. 2013). Once PAH-enriched particles accumulate in sediments, they may undergo a number of changes caused by chemical, biological, and physical activities. As a result, the bound PAHs can be remobilized from the sediment into the water phase and tend to bioaccumulate in aquatic organisms (Kumar et al. 2008). The PAHs are commonly determined by capillary gas chromatography, with flame ionization detection (GCFID), mass spectrometric detection (gas chromatography-mass spectrometry (GC-MS)), or by
Environ Monit Assess (2015) 187:474
high-performance liquid chromatography (HPLC) with ultraviolet (UV) or fluorescence detection (Martinez et al. 2004; Denis et al. 2012). The use of GC-MS for determination of PAHs is based on a favorable combination of greater selectivity, resolution, and sensitivity (Alvarez-Aviles et al. 2007; Yamada et al. 2009; Denis et al. 2012). However, due to the complex nature of the sediments, prior to introduction into the analytical system, a sample preparation (extraction and cleanup) process is needed (Shu et al. 2000; Alvarez-Aviles et al. 2007; Yamada et al. 2009). Sample extraction is a critical step in the determination of PAH because these compounds are strongly sorbed to the matrix; consequently, their extraction is time-consuming and, in many cases, causes quantification errors (Yamada et al. 2009). The extraction of PAHs from sediments can be achieved either with traditional or advanced methods. The traditional techniques require long extraction times, consume large amounts of solvent, and can degrade thermally labile compounds (Dean and Xiong 2000; Shu et al. 2003; Villar et al. 2004; Alvarez-Aviles et al. 2007; Wang et al. 2007; Chan et al. 2011). Currently, advanced extraction techniques, such as supercritical fluid extraction (SFE), accelerated solvent extraction (ASE), and microwave-assisted extraction (MAE), have been established in order to improve the drawbacks of traditional extraction techniques (Dean and Xiong 2000; Shu et al. 2000, 2003; Villar et al. 2004; Alvarez-Aviles et al. 2007; Wang et al. 2007; Nikolaou et al. 2009; Chan et al. 2011). The MAE is gaining an important role for sample preparation because it offers greatly reduced usage of organic solvents, reduced extraction times, and reduced size of extraction apparatus and permits simultaneous extraction of multiple samples (Dean and Xiong 2000; Villar et al. 2004; Wang et al. 2007; Chan et al. 2011). Two major rivers flow through Addis Ababa, Ethiopia, namely the Tinishu Akaki and Tiliku Akaki. These rivers carry domestic waste, rain runoff water, and untreated industrial discharges of various types, hence making them known for their offensive odor (Xu and Usher 2006; Prabu 2009). Lake Awassa and Ziway are found in the Ethiopian Rift Valley. The threats to the lakes are factory effluents, solid and liquid wastes generated by different sectors, small- and large-scale farming, and also sewer lines (Awulachew et al. 2007; Yohannes et al. 2013a, b). With regard to the status of potentially toxic elements, we recently reported that the sediments collected from the Akaki River was polluted
Environ Monit Assess (2015) 187:474
moderate or to a heavy degree by Cr, Zn, and Pb, notpolluted to moderate rate with Ni, and non-polluted by Cu (Mekonnen et al. 2014). Similarly, we also reported that Lake Awassa and Lake Ziway were not to moderately polluted with Cr and Ni but highly polluted with Zn (Mekonnen et al. 2015). However, no investigations have been carried out on the distribution and sources of PAHs in these important ecosystems. Yohannes et al. (2013a, b) investigated the levels of organochlorine pesticides (OCPs) and heavy metals in fish and sediment collected from Lake Awassa but not PAHs. Prasse et al. (2012) assessed contamination and sources of metals, polychlorinated biphenyles (PCBs), and PAHs in urban soils from Addis Ababa, Ethiopia, but not water systems. The objectives of this investigation were to assess the concentration, distribution, and sources of PAHs in sediments of the Akaki River, Lake Awassa, and Lake Ziway in Ethiopia using GC-MS after MAE and cleanup steps. In addition, ecological risks of these contaminants in surface sediments were also assessed by comparison with sediment quality guidelines (SQGs).
Experimental Chemicals and reagents For the extractions, 99.8 % acetone and 95 % n-hexane (HPLC grade, LAB-SCAN Analytical Sciences, Poland) were used. Copper metal powder was obtained from UNILAB (Philip Harris, UK) and activated using 99.9 % methanol (HPLC grade, LAB-SCAN Analytical Sciences, Poland), dichloromethane, and diethyl ether (AAR grade, SMM Instruments, South Africa). Anhydrous sodium sulfate (analyzed reagent >99.0 % UNIVAR) was obtained from Merck Chemicals (Pty) Ltd, Germany. Argon (99.998 % purity) and helium (99.999 %, purity) were supplied by Afrox Boc gases (Afrox, South Africa). The PAH kit containing 16 priority pollutants (Supelco Inc., Bellefonte, PA, USA) was used for preparation of series of standards for external calibration. A Standard Reference Material of New York-New Jersey Waterway Sediment (SRM 1944) (National Institute of Standards and Technology (NIST), Gaithersburg, MD, USA) was used to evaluate the accuracy and precision of the method. All the glassware and polyethylene bottles, prior to use, were washed with detergent, rinsed with doubly deionized water (18.2 M
Page 3 of 13 474
Ω/cm, Millipore purification system, Millipore, France), soaked in 10 % v/v HNO3 (65 % Suprapur®, Merck, Germany) for 24 h, and dried in an oven. Sample collection and preparation Three sampling areas, namely Addis Ababa, Awassa, and Ziway, were selected. From Addis Ababa, 15 sampling sites directly and/or indirectly related to Akaki River (AA Bulbula, AA Entoto, AA Gefersa, AA Kebena, AA Kera, AA Kidanemihret before TAR, AA TAR Alert, AA TAR before Kidanemihret, AA TAR before Melkaqurani, AA TAR Kaliti, AA Melkaqurani before TAR, AA TAR Kolfe, AA TAR with Kidanemihret, AA TlAR at KK Textile Factory, and AA TlAR with TAR at Aba Samuel) were selected for collecting sediment samples. Similarly, 11 sites from Awassa (Amoragedel, Asamaber, Buko, Fidelserawit, Kereisa, Mezinagna, Muate, Samijersa, Shewaber, Tikurwuha, and Wondotika) and ten sites from Ziway (Bochesa, Debresina, Gabriel, Gelila, Korekonch, Mezinagna, Shalo, Sedeicha, Wameicha, and Wofdeset) were selected (the maps of the sampling areas with the sampling points are attached as supplementary documents). Sediment samples were taken at a depth of 20 cm by a stainless steel grab sampler using 500-mL bottles which were kept in a cool container. Samples were air-dried, homogenized using an agate mortar and pestle and passed through a sieve (100 mesh), and stored until analysis. Microwave-assisted extraction A CEM MARS Xpress Microwave Accelerated Reaction System (CEM Corporation, Matthews, NC, USA) was used for extraction of PAHs. For extraction, EPA Method 3546 (microwave extraction) was used with some modifications and optimized for PAH extraction as follows. About 2-g sediment samples were weighed into extraction vessels, and extraction was done with the optimized method using 30 mL of mixture of n-hexane and acetone (1:1, v/v) for 30 min at a 110 °C at 200 W. After the vessels were allowed to cool to room temperature, the supernatant was filtered through a precleaned Pasteur’s pipette glass fitted with solvent-rinsed glass wool. Then, the solvent was decanted and subjected to a cleanup procedure (EPA Method 3660, sulfur cleanup). To the extract, 2 g of precleaned anhydrous Na2SO4 was added, mixed thoroughly, left overnight, and filtered. To
474
Environ Monit Assess (2015) 187:474
Page 4 of 13
the filtrate, 0.2 g of activated copper was added for desulfurization (Ledgard 2007), filtered with a premium syringe filter (diameter 0.22 μm, Agilent), and evaporated to nearly dryness using a Genevac EZ-2 Series Personal Evaporator (Genevac Inc, UK). To the concentrated extract, 1 mL of dichloromethane was added and transferred to 1.5-mL brown vials for GC-MS analysis.
The gas chromatography-mass spectrometry instrumentation and condition The PAHs in the purified extract were determined using an Agilent 7890A series gas chromatograph interfaced with an inert Agilent 5975C mass selective detector (MSD) with triple-axis detector (Agilent Technologies, USA) operated in electronic impact mode at 70 eV. Separation of the compounds was achieved using a fused silica capillary column (DB-5 MS, 30 m×0.25mm i.d.×0.25-μm film thickness; J&W Scientific, Folsom, CA, USA). The instrumental parameters are shown in Table 1.
Identification and quantification The identity of the PAHs in the sediment samples was confirmed by retention time and abundance of confirmation ions in the authentic PAHs standards. The USEPA 14 priority PAHs were quantified based on a minimum of seven-point calibration curve for individual compounds. Automated library searching was performed using the NIST Mass Spectral Database (Gaithersburg, MD, USA). The concentrations of PAHs are expressed on a dry mass basis.
Table 1 Operating conditions for GC-MS analysis Parameters
Description
Mobile phase
He
Column flow rate
0.8 mL/min
Oven temperature
Start at 60 °C and hold for 2 min, ramp to 160 °C at 15 °C/min and hold for 9 min, ramp to 260 °C at 5 °C/min and hold for 3 min, and then ramp to 300 °C at 5 °C/min and hold for 1.67 min with a total oven temperature run time of 50.34 min
MSD transfer line heater temperature Ion source temperature Quadrupole mass analyzer temperature Acquisition mode
310 °C 230 °C 150 °C
Identification in scan mode and quantification in selected ion monitoring (SIM) mode
Ions selected for each naphthalene (128, 127, 129, 102), PAH (m/z) acenaphthylene (152, 151, 76, 150), fluorene (166, 165, 167, 182), phenanthrene (178, 176, 179, 76), anthracene (178, 176, 179, 89), fluoranthene (202, 200, 203, 187), pyrene (202, 203, 101, 200), chrysene (228, 226, 229, 113), benzo[a]anthracene (228, 226, 229, 114), benzo[k]fluoranthene (252, 250, 253, 126), benzo[b]fluorantene (252, 253, 250, 126), benzo[a]pyrene (252, 253, 250, 126), indeno[1,2,3cd]pyrene (276, 277, 274, 136), and benzo[ghi]perylene (276, 137, 138, 274) The italicized QIons were used for quantification
Results and discussion Quality assurance
Optimization of microwave-assisted extraction
Analysis of the set of samples was accompanied by a procedural blank and a duplicate sample, which was analyzed in a similar fashion to the samples. The SRM 1944 was analyzed for quality assurance checks for the modified extraction EPA Method 3546 and cleanup procedure. Instrumental calibrations were checked by the injection of a continuing calibration solution (0.5 μg/ mL concentration). The GC-MS calibration was verified by running ten samples.
Microwave-assisted extraction In order to perform a simple, fast, and efficient extraction of PAHs, the amount of solvents, microwave power, and extraction time were studied. The literature indicated that the highest PAH recoveries were obtained using hexane/acetone (1:1, v/v) (Shu et al. 2000; Villar et al. 2004; Li et al. 2006; Svoboda et al. 2007; Wang et al. 2007). This solvent was therefore used for extraction.
Environ Monit Assess (2015) 187:474
Page 5 of 13 474
Sulfur is considered to be an interfering element in PAH analysis in sediment since it is soluble in organic solvents and co-extracted with target organic compounds. Removal of sulfur is possible using different reagents: potassium hydroxide in ethanol (Ricklund et al. 2008), tetrabutylammonium (TBA) hydrogen sulfate saturated with sodium sulfite (Svoboda et al. 2007), and activated copper (Ledgard 2007; Liang et al. 2007). Activated copper was therefore used for removal of sulfur from sediment samples. Method validation The relative standard deviation of duplicate samples was less than 10 %. The efficiency of MAE and the cleanup of sediment samples and sample blanks were checked using SRM 1944. The results indicated satisfactory recovery, with the average recovery ranging from 89.0 to 98.2 % (Table 2). The detection and quantification limits were set at three and ten times the background noise obtained for blank samples, respectively. The detection limit ranged from 1.28 ng/g (phenanthrene) to 3.92 ng/g (naphthalene), while the quantification limit ranged from Table 2 Certified and measured concentrations for selected PAHs in SRM 1944 for method validation PAHs
Certified concentration values (mg/kg dry mass)
Measured concentration (mg/kg dry mass, n=3)
Naphthalene
1.65±0.31
1.53±0.17
Acenaphthylene
NA
–
Fluorenea
0.85±0.03
0.77±0.06
Phenanthrene
5.27±0.22
5.11±0.16
Anthracene
1.77±0.33
1.63±0.19
Fluoranthene
8.92±0.32
8.53±0.27
Pyrene
9.70±0.42
8.88±0.41
Benz[a]anthracene
4.72±0.11
4.20±0.29
Chrysene
4.86±0.10
4.63±0.13
Benzo[b]fluoranthene
3.87±0.42
3.73±0.57
Benzo[k]fluoranthene
2.30±0.20
2.23±0.18
Benzo[a]pyrene
4.30±0.13
4.22±0.17
Indeno[1,2,3-cd]pyrene
2.78±0.10
2.73±0.15
Benzo[ghi]perylene
2.84±0.10
2.63±0.11
NA data not available a
This concentration is provided as a reference value
4.27 ng/g (phenanthrene) to 13.1 ng/g (naphthalene) (Table 3). The repeatability of at least two parameters in the gas chromatographic method: the retention time (confirming the identity of the analyte of interest) and the peak area or height (quantifying the analyte of interest) were achieved satisfactorily with the chromatographic method under the study conditions. Analytical determination The complete resolution of the 14 USEPA priority PAHs was obtained with a careful choice of suitable chromatography conditions, including the stationary phase, by carrying out a fast GC analysis on a DB-5 MS fused silica capillary column after MAE and cleanup. Quantification was performed using at least a seven-point calibration curve for each individual PAH. The correlation coefficient (R 2 ) varied from 0.955 to 0.998 (Table 3). However, for the two PAHs (acenaphthene and dibenzo(a,h)anthracene), the calibration curves were not sufficiently satisfactory to continue the work on the compound, and hence, only the remaining 14 PAHs were studied. The residual characteristics of PAH in studied area The total concentrations of PAHs (dry mass basis) are presented in Table 4. The highest total concentration of PAHs was found in the Akaki River (3070 ng/g) samples taken from at the TAR Alert area, while the lowest concentration ( phenanthrene (20 %) > fluoranthene (19 %) > pyrene (14 %) > anthracene (10 %) > fluorene (5 %) > benzo[ghi]perylene (2 %) ≈ acenaphthylene (2 %) > benzo[b]fluorantene (1 %) ≈ indeno[1,2,3-cd]pyrene (1 %) ≈ benzo[a]pyrene (1 %) ≈ chrysene (1 %) ≈ benzo[k]fluoranthene (1 %) > benzo[a]anthracene (0.3 %); the distribution varied substantially among the different locations. Therefore, of the 14 PAHs, naphthalene, phenanthrene, and fluoranthene were predominant species (accounting for 63 %) while benzo[a]pyrene,
474
Environ Monit Assess (2015) 187:474
Page 6 of 13
Table 3 List of PAHs for analysis, correlation coefficient, detection, and quantification limits for SIM GC-MS mode PAHs
Retention time, min
R2
Detection limit (ng/g, dm)
Quantification limit (ng/g, dm)
Chemical formula
No. of ring
Naphthalene
C10H8
2
7.869
0.994
3.92
13.1
Acenaphthalene
C12H8
3
10.998
0.995
1.46
4.87
Fluorene
C13H10
3
13.564
0.991
1.53
5.10
Phenanthrene
C14H10
3
20.106
0.983
1.28
4.27
Anthracene
C14H10
3
20.485
0.998
2.22
7.40
Fluoranthene
C16H10
4
27.515
0.993
2.77
9.23
Pyrene
C16H10
4
28.908
0.991
1.78
5.93
Benza[a]anthracene
C18H12
4
35.187
0.987
3.38
11.3
Chrysene
C18H12
4
35.007
0.978
3.38
11.3
Benzo[b]fluoranthene
C20H12
5
40.503
0.981
2.69
8.97
Benzo[k]fluoranthene
C20H12
5
40.656
0.976
2.69
8.97
Benzo[a]pyrene
C20H12
5
42.232
0.987
2.55
8.50
Indeno[1,2.3-cd]pyrene
C22H12
6
47.549
0.955
2.33
7.77
Benzo[ghi]perylene
C22H12
6
48.251
0.990
1.99
6.63
dm dry mass
chrysene, benzo[k]fluoranthene, and benzo[a]anthracene were found the least (accounting
