Investigating the Occurrence and Environmental Significance of

Sep 3, 2015 - ... Oxide and Inorganic Arsenic in Northern Hemisphere Summer Monsoonal Wet Deposition. Laurie Savage , Manus Carey , Mahmud Hossain , M...
0 downloads 10 Views 2MB Size
Download PDF
Article pubs.acs.org/est

Investigating the Occurrence and Environmental Significance of Methylated Arsenic Species in Atmospheric Particles by Overcoming Analytical Method Limitations Thrasyvoulos Tziaras, Spiros A. Pergantis,* and Euripides G. Stephanou Environmental Chemical Processes Laboratory, Department of Chemistry, University of Crete, Voutes Campus, Heraklion 71003, Greece S Supporting Information *

ABSTRACT: A novel analytical method has been developed for the determination of all five arsenic species known to exist in atmospheric particulate matter (PM), i.e., the inorganic arsenite iAs(III) and arsenate iAs(V), and the methylated methylarsonate (MA), dimethylarsinate (DMA) and trimethylarsine oxide (TMAO). Although the methylated species were first detected in PM in the late 1970s, most of the recent studies focus mainly on the two inorganic As species, ignoring TMAO in particular. In the present study, an HPLC (with an anion and cation exchange column connected in series)-arsine generation-ICP-MS system provided complete separation of all five As species and limits of detection from 10 to 25 pg As mL−1. This method was applied to analyze water extracts of the inhalable fraction of atmospheric PM (PM10, PM2.5 and PM2.1). 81 samples were collected, most during Saharan dust events, from a semirural area, and analyzed. The total water extractable arsenic ranged from 0.03 to 0.7 ng of As m−3, values that are representative for remote areas. iAs(V) was the most abundant species followed by TMAO, DMA, iAs(III) and MA. None of the As species showed any particular trend with the presence or intensity of dust events, or seasonality, except for TMAO, which showed higher concentrations during the colder months.

■

INTRODUCTION Since the 1989 comprehensive review on arsenic (As) speciation in the environment by Cullen and Reimer,1 our knowledge on As species behavior in the atmosphere and their role on the global As budget has failed to progress to any significant extent, especially regarding methylated As species. One of the earliest studies on As speciation in the atmosphere, referenced in the review, was by Johnson and Braman2 on alkyland inorganic As in air samples and their significance on the global arsenic budget. In this study, inorganic arsenic, dimethylarsine and trimethylarsine species were detected and quantitated in a variety of urban, suburban and rural air samples. Approximately 20% of the total As (1.7 ng As m−3) found in these air samples was in the alkyl-arsenic form. Also mentioned in the review were studies by Mukai at al.3,4 demonstrating the seasonal variation of the methylarsenic species concentrations in airborne particulate matter. They reported that for two Japanese sites the ratio of the di- to trimethyl arsenic species was usually between 0.15 and 0.34, indicating that (CH3)3As was the dominating methylated species thought to have been formed via biomethylation. Also, the methylarsenic species concentrations were found to be higher in the summer than in the winter, in accordance with the observed temperature changes. For the trace amounts of monomethylated species, detected in the summer, they attributed their presence to contamination by a methylarsonate pesticide used near the sampling site. In addition, the © 2015 American Chemical Society

determined arsenic concentrations indicate a significant enrichment over crustal arsenic concentrations, thus providing strong support for a significant biomethylation contribution to the atmospheric arsenic cycle. Briefly, we now discuss what has been published in the years following the 1989 review.1 First of all, a major analytical development was the introduction of inductively coupled plasma mass spectrometry (ICP-MS) and its hyphenation to liquid chromatography (LC) and gas chromatography (GC). This has enabled improved selectivity, sensitivity and lower limits of detection, thus providing significantly improved capabilities for As speciation analysis. The use of hyphenated LC-ICP-MS methods resulted in extensive breakthroughs in As speciation analysis for numerous environmental and biological systems.5 However, as species identification is based on matching the retention times determined for arsenic standards with those of the sample peaks, certain precautions need to be taken as described in a recent critical review.6 It is of interest to note that in several recent arsenic speciation in atmospheric particles studies the use of LC-hydride generation (HG)-atomic fluorescence spectrometry (AFS)7−10 and LC-ICP-MS11 based methods did not reveal the presence of trimethylated arsenic Received: Revised: Accepted: Published: 11640

May 10, 2015 August 29, 2015 September 3, 2015 September 3, 2015 DOI: 10.1021/acs.est.5b02328 Environ. Sci. Technol. 2015, 49, 11640−11648

Article

Environmental Science & Technology

Figure 1. Dual column (anion + cation exchange) HPLC connected to an online arsine generation system for the gas phase introduction of arsines into the ICP-MS. A slightly modified conical spray chamber was used as the gas−liquid separator.

as a complementary approach to anion exchange ICP-MS to confirm the presence of TMAO in PM10 samples.13 However, even in the presence of such an excellent study, further confusion is added by a subsequent 2012 review12 on arsenic speciation in atmospheric particles that fails to reference the 2010 Jakob et al. study.13 Even more troubling is that it states that because TMAO (referred to as trimethylarsenate in the review) is only mentioned in the two Mukai studies3,4 it is not discussed further in the review. So, clearly there is an urgent need for clarification with respect to which arsenic species are actually present in atmospheric particles, as well as demonstrating a suitable analytical approach for their speciation. In view of the current state of the literature on As speciation in atmospheric particles, the objective of the present study was to develop a chromatographic approach for the complete separation of all As species that have been reported to date to be present in atmospheric particulate matter, i.e., iAs(III), iAs(V), MA, DMA and TMAO. To achieve this in a single separation, an anion exchange column connected in-series with a cation exchange column were used. Additionally, for improved sensitivity and limits of detection, a postcolumn arsine generation (AG) system was used along with ICP-MS detection. The developed dual column HPLC-AG-ICP-MS system was used to analyze 81 water extracts from atmospheric particle samples of various sizes, i.e., PM2.1, PM2.5 and PM10, with an emphasis on samples collected from a semirural area during Saharan dust events, which are frequent in the vicinity of the Greek island of Crete situated in the Eastern Mediterranean. For comparison purposes, samples were also collected before and after such events. The study emphasizes the need to accurately identify and quantify the As species present in atmospheric particles in order to lay the foundations for a better understanding of their sources, how they are produced and whether or not such processes may be of environmental significance.

species, i.e., trimethylarsine or trimethylarsine oxide (TMAO), which were clearly shown to be present in the 1970 and 1980 studies.2−4 In contrast, inorganic arsenic in the +III oxidation state (iAs(III)) was reported with an increased frequency in almost all the contemporary studies. A detailed list of studies failing to identify TMAO, but having detected iAs(III) and other As species, are given in a recent review article (Table 1 in ref 12).12 This discrepancy makes it increasingly difficult to understand As species behavior and their origin in the atmosphere. From a methodological point of view, most of the contemporary As speciation studies use an anion exchange column (Hamilton PRP-X100). Even though this is an extremely efficient chromatographic column for As speciation analysis and one of the most frequently used, it does not retain iAs(III) to any substantial extent, and as shown by Jakob et al., it does not retain TMAO either.13 Thus, when an As-containing peak is observed in the column’s void volume, it is not possible to identify if it is iAs(III) and/or TMAO. Thus, without any additional chromatography or intentional species conversion (e.g., oxidation to convert iAs(III) to iAs(V)), it is not possible to identify definitively the nonretained peak eluting from an anion exchange Hamilton PRP-X100 column as this may contain either iAs(III) or TMAO or a mixture of both. All contemporary studies,7−9,11,14 with one exception,13 have failed to mention this shortcoming in their applied analytical approach, thus not clarifying if any attempt was made to address it. Therefore, currently, an unsatisfactory situation exists in which numerous publications using HPLC-ICP-MS have failed to identify the presence of trimethylarsenic species in atmospheric PM, whereas they report on the presence of iAs(III) using an analytical approach that is not suitable to differentiate between iAs(III) and TMAO. Because of this, it may be necessary for these results to be re-examined and clarified and a sound analytical methodology be developed and used. The one exception is a recent study by Jakob et al., who demonstrate excellent analytical chemistry, recognizing the previously mentioned analytical shortcoming of the anion exchange ICP-MS approach.13 To overcome it, they use cation exchange chromatography and electrospray mass spectrometry

■

EXPERIMENTAL SECTION Instrumentation and Equipment. Two inductively coupled plasma mass spectrometers were used in this study, i.e., an X-series ICP-MS (Thermo Fischer Scientific, Winsford, 11641

DOI: 10.1021/acs.est.5b02328 Environ. Sci. Technol. 2015, 49, 11640−11648

Article

Environmental Science & Technology Table 1. Sampling Informationa sampling period

sample ID

sampling characterization

sampler type

27/09/2012−11/03/2013 21/05/2013−30/05/2013 22/07/2013−27/07/2013 15/10/2013−21/01/2014 17/02/2014−27/02/2014 27/03/2014−20/11/2014 15/09/2014 06/11/2014

S1−S14 S19−S22 S23−S28 S29−S32 S33−S38 S39−S53 S54 S55

D.E. D.E. seasonal−summer D.E. seasonal−winter 11, D.E.; 3, N.SD. N.SD. N.SD.

H.V. L.V. L.V. L.V. L.V. L.V. H.V. H.V.

size fractionation (n: number of samples) PM2.1 (13) PM2.5 (4) PM2.5 (6) PM2.5 (4) PM2.5 (5) PM2.5 (14) PM10 (1) TSP (1)

PM10 PM10 PM10 PM10 PM10

(4) (6) (4) (5) (14)

a

D.E. stands for Saharan dust events, N.S.D. for no trace of transferred Saharan dust, H.V. and L.V. for high and low volume samplers respectively, TSP for total suspended particles.

and in the SI Sampling Details section. In total, 81 samples were collected and analyzed, including 13 PM2.1, 1 PM10 and 1 TSP high volume samples and 33 PM2.5 and another 33 PM10 low volume samples. Focusing mainly on particulate matter transferred by air masses coming from Africa’s deserts, air monitoring usually took place during days with strong or weak dust outbreak events. Moreover, for comparison purposes, some samples were taken before and after the dust events. To study seasonal variations, samples were collected during both winter and summer month sampling campaigns (Table 1). Seventy two (72) h backward trajectories were calculated for each sample by using the HYSPLIT model18 (representative plots are presented in Figure S1). After each sampling, the resulting filters were wrapped in aluminum foil and kept in airtight zip-lock plastic bags at −26 °C until extraction or digestion. Prior to sampling, all quartz filters and aluminum foils had been heated to 550 °C for 4 h in order to remove any existing contaminants. Sample Extraction. For sample extraction, pieces of filter corresponding to a portion of 2.5% of the high volume filters and of 43% of the low volume filters were cut and placed into 50 mL sample tubes and extracted using 2.0 or 1.5 mL of deionized water, respectively. The sample tubes were sealed tightly and ultrasonicated for 90 min at 50−55 °C. Afterward, the extracts were filtered through 0.45 μm syringe filters and kept in capped 1.5 mL microtubes, which were stored at 4 °C and analyzed within 48 h. The extraction efficiency was evaluated as follows, first the total As was determined following microwave acid digestion of the filter samples, subsequently the determined total As was compared to the sum of the concentrations of the As species determined by using HPLC-AG-ICP-MS to be present in the water extracts. Thirteen (13) high volume filter samples (11 × PM2.1, 1 × PM 10 and 1 × TSP) were analyzed for both total As and As species content. The recoveries mentioned in the Results and Discussion section are the combined PM extraction efficiencies and HPLC column recoveries, as the two were not determined separately during the arsenic speciation analysis. For total arsenic determination, a section (7.20% of the entire filter) of a high volume filter was put into a 60 mL Teflon vessel and digested with 5 mL of 69−70% HNO3. This was achieved using a two-step microwave digestion program: step 1 was at 180 °C using 990 W of power for 60 min, and step 2 was set at 100 °C using 800 W of power for 20 min. The digest was left to cool and was then diluted with deionized water 14-fold in order to reduce the concentration of nitric acid to 5% w/v and to be analyzed using ICP-MS. The concentration of Mn

UK) and a NexION 300-XX ICP-MS (PerkinElmer). The same conical spray chamber with an impact bead (originally supplied with the X-series ICP-MS instrument) was used with both instruments. Arsenic separation was achieved using a LC-20AD Prominence gradient pump (Shimadzu, Kyoto, Japan). Arsine Generation System Built in-House for ICP-MS. The arsine generation (AG) system, located postcolumn, was used to convert the arsenic species eluting from the HPLC column to their corresponding arsines and efficiently transporting them into the ICP-MS (Figure 1). This was achieved by connecting the HPLC column outlet to two in-series T-piece mixers. The first was used to acidify the HPLC eluent by adding 0.25% w/v HCl, at a flow rate of 1.4 mL min−1, whereas the second T-piece mixer was used to introduce the reducing agent (1% w/v NaBH4 at a flow rate of 1.1 mL min−1). A peristaltic pump was used to deliver all additional flows as well as for waste removal. Efficient arsine generation was achieved in a 1 m knotted reaction coil which went directly into the conical spray chamber via a hole that had been drilled through the Teflon waste plug. This allowed for the spray chamber to function as a gas/liquid separator. Argon gas, provided via the nebulizer, flushed the generated arsines from the spray chamber into the ICP-MS. It should be noted that the nebulizer had no other function apart from providing the necessary Ar gas required to flush the arsines and carry them efficiently into the plasma. It should be stressed that waste removal was critical, as the total waste generated was high (approximately 3 mL min−1) and its efficient removal had to be conducted in a nonpulsating continuous fashion. Chemicals and Materials. Standard solutions used for total arsenic determination were prepared daily from a single element standard stock solution of 10 000 μg As mL−1 from CPI international. Aqueous solutions containing individual arsenic compounds were prepared from sodium arsenite, NaAsO2 (BDH Chemicals Ltd., Poole, England), from sodium arsenate dibasic heptahydrate Na2HAsO4·7H2O and dimethylarsinic acid or cacodylic acid (Fluka) and from monosodium acid methane arsonate sesquihydrate (Chem Service, Wester Chester, PA). Trimethylarsine oxide (CH3)3AsO had been prepared by literature methods,15 and has been characterized by using mass spectrometry in previous studies.16,17 Sample Collection. Twenty four (24) h sampling sessions were conducted on the northern front of the Department of Chemistry, University of Crete, (35°18′ N, 25°45′ E) located in a semirural site approximately 6 km SW of the city of Heraklion, Crete, and about the same distance from a petrol powered electricity station which is located NW of the University of Crete. Sampling details are included in Table 1 11642

DOI: 10.1021/acs.est.5b02328 Environ. Sci. Technol. 2015, 49, 11640−11648

Article

Environmental Science & Technology

Figure 2. Anion exchange ICP-MS chromatogram pairs corresponding to the arsenic species before and after the addition of hydrogen peroxide for (a) atmospheric particle extracts, (b) iAs(III) standard and (c) TMAO standard. In this case, HPLC eluent was introduced into the ICP-MS using pneumatic nebulization.

protocol described previously. Some addition extraction procedures were also evaluated (conditions given in Table S1) in order to improve further extraction efficiency. To recover nonextracted arsenic species remaining from the first extraction, a second extraction with hydrogen peroxide was performed on the same filters after the water extracts were removed. Concentrated hydrogen peroxide (1.5 mL) was

and other elements, besides As, was determined using multielemental ICP-MS analysis in order to investigate the intensity of the Saharan dust events and the magnitude of the mineral content of PM. For sample extraction, a piece of the high volume filters corresponding to a portion of 1.44% of the entire filter were cut and extracted using 1.5 mL of deionized water following the 11643

DOI: 10.1021/acs.est.5b02328 Environ. Sci. Technol. 2015, 49, 11640−11648

Article

Environmental Science & Technology

Figure 3. Dual column LC-ICP-MS chromatograms for (a) a standard solution containing 5 arsenic species at 0.2 μg L−1 As for each species, (b) an atmospheric PM2.5 water extract and (c) an atmospheric PM10 water extract were all obtained using HPLC coupled online with ICP-MS using an arsine generating interface. The (a) chromatogram was offset by +14 000 counts per second (cps) for better viewing, whereas the (b) chromatogram by +7000 cps. The sample extract corresponding to trace b was determined to contain As at 3.6 pg m−3 for Aite, 3.4 pg m−3 for DMA, 42.0 pg m−3 for TMAO, MA was not detected and 249.9 pg m−3 for iAs(V), whereas trace c showed As at 2.2 pg m−3 for Aite, 5.3 pg m−3 for DMA, 31.1 pg m−3 for TMAO, traces of MA and 234.4 pg m−3 for iAs(V).

added to 13 samples (11 × PM2.1, 1 × PM10 and 1 × TSP) and sonicated for another 90 min at 50−55 °C. These extracts were analyzed for their As species content, taking into account that iAs(III) is oxidized to iAs(V) under such conditions. As Speciation Analysis. The identification of the arsenic species in the atmospheric particle matter extracts was based on matching their chromatographic peak retention times with those of arsenic reference standards. The arsenic species detected by using HPLC-AG-ICP-MS were quantified by using external calibration in the concentration range from 0.08 to 0.8 ng As mL−1 for each of the five As species expected to be present, i.e., iAs(III), DMA, TMAO, MA and iAs(V). Thioarsenic Investigation. It is well-known that thioarsenic species are present in marine organisms,19−22 and thus their occurrence in particulate matter also needed to be investigated. Because the thioarsenic species are expected to be irreversibly retained by the chromatographic conditions used, hydrogen peroxide was added to oxidize them to their corresponding oxo-analogues. Therefore, 50 μL of hydrogen peroxide was added to nine water extracts (0.5 mL). The H2O2 treated samples were allowed to stand for 30 min and then reanalyzed using HPLC-AG-ICP-MS. The determined As species were compared to those determined in the original water extracts.

■

to the water extract sample. It was expected that if the nonretained peak was iAs(III), it would be oxidized to iAs(V), which is what is observed for standard iAs(III) upon addition of H2O2 (Figure 2b). However, when H2O2 was added to the atmospheric particulate matter extract, no species conversion was observed, strongly suggesting that the nonretained peak shown in Figure 2a was not iAs(III). In a similar experiment, H2O2 added to a standard solution containing TMAO did not cause any conversion (Figure 2c). This was expected because As in TMAO is already in its +V oxidation state. The results from these experiments indirectly suggest that the nonretained species in the extract chromatogram could be TMAO. However, such a speculative identification required further confirmation. To obtain additional confirmation, we introduced a cation exchange column in-series after the anion exchange HPLC column. This did not provide any additional retention for the four As species iAs(III), DMA, MA and iAs(V), but did provide added retention for TMAO (pKa = 3.6) using a mobile phase consisting of 10 mM (NH4)2HPO4 at pH 4.2. These conditions allowed for the complete separation of all five As species using the dual column configuration (Figure 3a−c). It is evident that iAs(III) is not retained by any of the two columns, whereas TMAO, being partially cationic at pH 4.2, is retained considerably by the cation exchange column and can thus be separated from the other four As species. This is an important improvement in the analytical methodology for As speciation in atmospheric particle extracts, as it allows for complete separation of all five of the As species reported to be present in such particles. Most notable is the fact that the complete separation all As species is now possible in a single chromatographic analysis. Because sensitivity remained an issue, the pneumatic nebulizer commonly used for sample introduction into the ICP-MS was replaced by an arsine generation (AG) system. This allowed for more efficient sample introduction, thus higher sensitivity and potential for improved limits of detection. The arsine generation approach is commonly referred to as hydride generation in the literature; however, this term is avoided in the present study because the main species detected, i.e., TMAO, is converted to trimethylarsine (CH3)3As, which is not a hydride. An in-house constructed arsine generator was

RESULTS AND DISCUSSION

As Separation and Detection. Our initial attempt to conduct As speciation analysis in atmospheric particle extracts involved using an analytical approach similar to what has been described in most of the recently published studies. This involves using an anion exchange column for the separation of four As compounds, i.e., iAs(III), DMA, MA and iAs(V). The resulting chromatogram from atmospheric particle extracts revealed As-containing peaks that had retention times matching those of iAs(V) and DMA (minor peak). In addition, a high intensity peak corresponding to a nonretained As species was also observed (Figure 2a). Initially, it was suggested that this peak corresponds to iAs(III). However, the studies of Johnson and Braman,2 Mukai3,4 and the Jakob et al.13 prompted us to investigate further for the presence of TMAO, which is not retained under the applied chromatographic conditions. This was conducted by adding a small amount of concentrated H2O2 11644

DOI: 10.1021/acs.est.5b02328 Environ. Sci. Technol. 2015, 49, 11640−11648

Article

Environmental Science & Technology

Figure 4. Seasonal variation of TMAO in PM2.5 and PM10 samples.

samples with values of 76 ± 16%, 20 ± 15%, 2 ± 1%, 2 ± 1% for iAs(V), TMAO, DMA and iAs(III), respectively. MA was detected in 6 PM10 samples at relatively low concentrations of a few pg As m−3. A significant correlation (R2 = 0.84, p < 0.05) was found between the sum of the As species found in PM2.5 and PM10 (Figure S4). However, slightly higher concentrations were found for the PM10 samples, as indicated by the slope of the correlation regression line, i.e., 1.15. This indicates that approximately 88% of the water extractible arsenic is in the 300 pg As m−3 during two successive summers.3 In another of their studies,27 they determined MA in PM to be present at 1.4 ng As m−3. In both studies, they attributed the high levels of MA to contamination by the pesticide iron methane arsonate used in nearby rice fields. Extraction Efficiency. The high volume TSP sample S55 (Table 1) was chosen for extraction efficiency experiments due to a more homogeneous distribution of the particles across the filter compared to fractionated samples. Three pieces of the filter were acid digested, and total As was determined in all three showing a low relative standard deviation of 4.1%. Various aqueous based solutions were used for the filter treatment/ extraction and the results are presented in Table S1. Most of the applied media showed sufficient recoveries and the ultrasonication with deionized water for 90 min at 50 °C heating was chosen as the extraction procedure, as it provides acceptable extraction efficiency for each of the five individual As species and prevents the introduction of large amounts of the extracting agents (100 μL injections) onto the dual column AG-ICP-MS system. Similar results were reported by Oliveira et al., who applied similar extraction methods.32 Thirteen of the high volume samples were subjected to total arsenic and manganese determination and to arsenic speciation of their initial water extract and of a second subsequent hydrogen peroxide extract in order to investigate for more tightly bound As contained in the particulate matter. Details about sample characteristics and their total arsenic species content are included in Table S2. Eleven of the samples were PM2.1 and influenced by African dust transfer to various extents and two samples (one PM10 and one TSP) were regarded as free of African dust. The total As varied between 0.17 and 0.84 ng·m−3 and the total Mn from 1.07 to 64.08 ng·m−3 revealing a

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.5b02328. Equipment, Chromatographic Separation and Acid Digestion Chemicals, Sampling Details and Calculation of Limits of Detection sections. Methods of As extraction (Table S1), arsenic species recoveries (Table S2), contribution of each arsenic species to the total recovered As (Table S3), backward trajectories (Figure S1), concentrations of iAs(III) and iAs(V) (Figure S2), concentrations of DMA and TMAO (Figure S3), correlation of total As (sum of all determined As species) determined in PM2.5 and PM10 water extracts (Figure S4) and correlation of As in PM2.5 and PM10 in water extracts for TMAO (Figure S5) (PDF).

■

AUTHOR INFORMATION

Corresponding Author

*S. A. Pergantis. E-mail: [email protected]. Tel.: +30 2810 545084. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The work part of the project “CHEMISAND” is implemented under the “ARISTEIA” Action of the “OPERATIONAL PROGRAMME EDUCATION AND LIFELONG LEARNING” and is cofunded by the European Social Fund (ESF) and National Resources.

■

REFERENCES

(1) Cullen, W. R.; Reimer, K. J. Arsenic Speciation in the Environment. Chem. Rev. 1989, 89, 713−764.

11647

DOI: 10.1021/acs.est.5b02328 Environ. Sci. Technol. 2015, 49, 11640−11648

Article

Environmental Science & Technology (2) Johnson, D. L.; Braman, R. S. Alkyl- and Inorganic Arsenic in Air Samples. Chemosphere 1975, 4, 333−338. (3) Mukai, H.; Ambe, Y.; Muku, T.; Takeshita, K.; Fukuma, T. Seasonal variation of methylarsenic compounds in airborne particulate matter. Nature 1986, 324, 239−241. (4) Mukai, H.; Ambe, Y. Determination of Methylarsenic Compounds in Airborne Particulate Matter by Gas Chromatography with Atomic Absorption Spectrometry. Anal. Chim. Acta 1987, 193, 219−229. (5) B’Hymer, C.; Caruso, J. A. Arsenic and its speciation analysis using high-performance liquid chromatography and inductively coupled plasma mass spectrometry. J. Chromatogr. A 2004, 1045, 1− 13. (6) Francesconi, K. A.; Kuehnelt, D. Determination of arsenic species: a critical review of methods and applications, 2000−2003. Analyst 2004, 129, 373−395. (7) Farinha, M. M.; Slejkovec, Z.; van Elteren, J. T.; Freitas, M. C. Arsenic Speciation in Lichens and in Coarse and Fine Airborne Particulate Matter by HPLC−UV−HG−AFS. J. Atmos. Chem. 2004, 49, 343−353. (8) Sánchez de la Campa, A. M.; de la Rosa, J. D.; Sánchez-Rodas, D.; Oliveira, V.; Alastuey, A.; Querol, X.; Gómez Ariza, J. L. Arsenic speciation study of PM2.5 in an urban area near a copper smelter. Atmos. Environ. 2008, 42, 6487−6495. (9) Sánchez-Rodas, D.; Sánchez de la Campa, A. M.; de la Rosa, J. D.; Oliveira, V.; Gómez-Ariza, J. L.; Querol, X.; Alastuey, A. Arsenic speciation of atmospheric particulate matter (PM10) in an industrialised urban site in southwestern Spain. Chemosphere 2007, 66, 1485−1493. (10) Yang, G.; Ma, L.; Xu, D.; Li, J.; He, T.; Liu, L.; Jia, H.; Zhang, Y.; Chen, Y.; Chai, Z. Levels and speciation of arsenic in the atmosphere in Beijing, China. Chemosphere 2012, 87, 845−850. (11) Huang, M.; Chen, X.; Zhao, Y.; Yu Chan, C.; Wang, W.; Wang, X.; Wong, M. H. Arsenic speciation in total contents and bioaccessible fractions in atmospheric particles related to human intakes. Environ. Pollut. 2014, 188, 37−44. (12) Lewis, A. S.; Reid, K. R.; Pollock, M. C.; Campleman, S. L. Speciated arsenic in air: Measurement methodology and risk assessment considerations. J. Air Waste Manage. Assoc. 2012, 62, 2−17. (13) Jakob, R.; Roth, A.; Haas, K.; Krupp, E. M.; Raab, A.; Smichowski, P.; Gómez, D.; Feldmann, J. Atmospheric stability of arsines and the determination of their oxidative products in atmospheric aerosols (PM10): evidence of the widespread phenomena of biovolatilization of arsenic. J. Environ. Monit. 2010, 12, 409−416. (14) González-Castanedo, Y.; Sanchez-Rodas, D.; Sánchez de la Campa, a. M.; Pandolfi, M.; Alastuey, a.; Cachorro, V. E.; Querol, X.; de la Rosa, J. D. Arsenic species in atmospheric particulate matter as tracer of the air quality of Doñana Natural Park (SW Spain). Chemosphere 2015, 119, 1296−1303. (15) Kaise, T.; Hanaoka, K.; Shoji, T. The formation of trimethylarsine oxide from Arsenobetaine by Biodegradation with Marine Microorganisms. Chemosphere 1987, 16, 2551−2558. (16) Nischwitz, V.; Pergantis, S. A. Optimisation of an HPLC Selected Reaction Monitoring Electrospray Tandem Mass Spectrometry Method for the Detection of 50 Arsenic Species. J. Anal. At. Spectrom. 2006, 21, 1277−1286. (17) Nischwitz, V.; Pergantis, S. A. Liquid Chromatography Online with Selected Reaction Monitoring Electrospray Mass Spectrometry for the Determination of Organoarsenic Species in Crude Extracts of Marine Reference Materials. Anal. Chem. 2005, 77, 5551−5563. (18) G.D, D. R. R.; H. An overview of the HYSPLIT_4 modelling system for trajectories. Aust. Met. Mag. 1998, 47, 295−308. (19) Fricke, M. W.; Parks, A. N.; Schwegel, C. A.; Creed, J. T. Extraction and detection of a new arsine sulfide containing arsenosugar in molluscs by IC-ICP-MS and IC-ESI-MS/MS. J. Anal. At. Spectrom. 2004, 19, 1454−1459. (20) Maher, W. A.; Foster, S.; Krikowa, F.; Duncan, E.; St John, A.; Hug, K.; Moreau, J. W. Thio arsenic species measurements in marine organisms and geothermal waters. Microchem. J. 2013, 111, 82−90.

(21) Nischwitz, V.; Kanaki, K.; Pergantis, S. A. Mass spectrometric identification of novel arsinothioyl-sugars in marine bivalves and algae. J. Anal. At. Spectrom. 2006, 21, 33−40. (22) Soeroes, C.; Goessler, W.; Francesconi, K. A.; Schmeisser, E.; Raml, R.; Kienzl, N.; Kahn, M.; Fodor, P.; Kuehnelt, D. Thio arsenosugars in freshwater mussels from the Danube in Hungary. J. Environ. Monit. 2005, 7, 688−692. (23) Anderson, S. L.; Pergantis, S. A. Sequential hydride generation/ pneumatic nebulisation inductively coupled plasma mass spectrometry for the fractionation of arsenic and selenium species. Talanta 2003, 60, 821−830. (24) World Health Organisation. Air Quality Guidelines for Europe, 2nd ed.; WHO Reg. Publ. Reg. Off. Eur.: Copenhagen, Denmark, 2001. (25) Schroeder, W. H.; Dobson, M.; Kane, D. M.; Johnson, N. D. Toxic Trace Elements Associated with Airborne Particulate Matter: A Review. JAPCA 1987, 37, 1267−1285. (26) Walsh, P. R.; Duce, R. A.; Fasching, J. L. Tropospheric arsenic over marine and continental regions. J. Geophys. Res. 1979, 84 (NC4), 1719−1726. (27) Mukai, H.; Ambe, Y. Detection of monomethylarsenic compounds originating from pesticide in airborne particulate matter sampled in an agricultural area in Japan. Atmos. Environ. 1987, 21, 185−189. (28) Tsopelas, F.; Tsakanika, L.-A.; Ochsenkühn-Petropoulou, M. Extraction of arsenic species from airborne particulate filters Application to an industrial area of Greece. Microchem. J. 2008, 89, 165−170. (29) EU. 2004/107/CE Council Directive relating to arsenic, cadmium, mercury, nickel and polycyclic aromatic hydrocarbons in ambient air. Off. J. Eur. Union, L: Legis. (Eng. Ed.) 2004. (30) Tsapakis, M.; Stephanou, E. G. Polycyclic aromatic hydrocarbons in the atmosphere of the eastern Mediterranean. Environ. Sci. Technol. 2005, 39, 6584−6590. (31) Kostandoulaki, A. Arsenic in Farmers’ Food. Patris (Greek Newspaper) (http://www.patris.gr/articles/151425/89715#. VRLonPmsW7M), 2009. (32) Oliveira, V.; Gómez-Ariza, J. L.; Sánchez-Rodas, D. Extraction procedures for chemical speciation of arsenic in atmospheric total suspended particles. Anal. Bioanal. Chem. 2005, 382, 335−340. (33) Gamaletsos, P.; Godelitsas, A.; Dotsika, E.; Tzamos, E.; Gottlicher, J.; Filippidis, A.; Scozzari, A.; Dotsika, E. Geological Sources of As in the Environment of Greece: A Review. In Handbook of Environmental Chemistry; Scozzari, A., Dotsika, E., Eds.; Springer: New York, 2006; Vol. 5, pp 1−12. (34) Matschullat, J. Arsenic in the geosphere-a review. Sci. Total Environ. 2000, 249, 297−312.

11648

DOI: 10.1021/acs.est.5b02328 Environ. Sci. Technol. 2015, 49, 11640−11648