Evaluation of commonly used filter substrates for the measurement of aerosol trace element solubility

The published literature describing aerosol trace element fractional solubility measurements is characterized by a wide range of observed fractional solubilities. Whereas some of this variability is derived from natural differences in the chemical characteristics of the aerosol source material, the use of different sample collection and processing protocols by the scientific community has also confounded efforts to understand aerosol solubility. Bulk aerosol samples were collected at a coastal site over a nine-month period and used to assess the influence of filter material on aerosol solubility measurements. Two hundred samples were extracted with ultrapure deionized water and focused on the solubility of ten trace elements (Al, P, Ti, V, Mn, Fe, Ni, Cu, Zn, and Pb) of interest to the GEOTRACES program. Aerosol samples were collected on eight different filter types and extracted using a flow-through “instantaneous” extraction method. In many cases, the operationally defined aerosol trace element solubility differed depending on filter type. Major anion concentrations and trace element fractional solubility were found to differ 58% and 60% of the time, respectively. Filter blank concentrations are also reported for the various filter types. This work, in conjunction with the 2008 GEOTRACES Aerosol Intercalibration study, should aid the design of future research efforts by the wider marine aerosol community and allow better comparisons among published data. *Corresponding author: E-mail: [email protected]; Earth, Ocean, and Atmospheric Sciences, Florida State University, Tallahassee, FL 32306; Phone: 850.644.0259 Acknowledgments This work was supported by NSF grant OCE 0824304 to A.P. We thank Mark Chernihovsky (IUI, Israel) for coordinating aerosol collection in Israel. We also wish to thank Rob Franks (UCSC) for generously lending his expertise and support during ICP-MS method development. The authors gratefully acknowledge the NOAA Air Resources Laboratory (ARL) for the provision of the HYSPLIT transport and dispersion model and READY website (http://www.arl.noaa.gov/ready.html) used in this publication. Dr. Peter Croot and two anonymous reviewers were instrumental in improving this manuscript. This work is part of the Intercalibration in Chemical Oceanography special issue of L&O Methods that was supported by funding from the U.S. National Science Foundation, Chemical Oceanography Program. DOI 10.4319/lom.2012.10.790 Limnol. Oceanogr.: Methods 10, 2012, 790–806 © 2012, by the American Society of Limnology and Oceanography, Inc. LIMNOLOGY and OCEANOGRAPHY: METHODS cantly impacted by the deposition of anthropogenic aerosols (Sholkovitz et al. 2010). Laboratory experimentation on both natural samples and reference materials using a variety of extraction methods has resulted in a wide range of operationally defined aerosol Fe fractional solubilities (Mahowald et al. 2005; Mahowald et al. 2008). We spotlight Fe as an example due to the relatively large number of published aerosol Fe solubility datasets, however it is likely that similar solubility ranges exist for many other elements (e.g., P). Baker and Croot (2010) suggest, “...some of this variability in reported estimates must also be due to the diverse experimental approaches used for determination of aerosol iron solubility.” These approaches may include the use of different extraction solutions, solvent pH, solution to particle ratio, and exposure time with each of these variables potentially influencing the operationally defined fractional solubility estimate. In their review, those authors go on to rightly put forth surface seawater as the most “obvious” extraction media for assessing aerosol solubility in the ocean. However, as they point out and as is discussed in Buck et al. (2010) and Sholkovitz et al. (2012), the use of seawater complicates interpretation of the results and increases the complexity of subsequent analyses. Recent research has striven to offer alternative extraction solutions, and several studies have compared the solubility of aerosol Fe in both 18.2 mW ultrapure deionized water and 0.2 μm filtered surface seawater both from marine and coastal aerosols (Buck et al. 2006; Chen et al. 2006; Aguilar-Islas et al. 2010; Buck et al. 2010). In one case, the fractional solubility was greater in seawater than in UHP (Aguilar-Islas et al. 2010). As noted in Aguilar-Islas et al. (2010), it is possible that experimental artifacts are less likely to influence observational variability than aerosol source and together these factors explain the range of published aerosol Fe fractional solubilities in the literature. Regardless of the source of the variability, this range in solubilities complicates the effort to produce a widely applicable estimate of aerosol solubility, limits study comparability, and ultimately hinders our understanding of the factors controlling solubility. One of the specific GEOTRACES objectives is to “establish the range of fractional solubility of key atmospheric components and the process that underlie that variability” (SCOR Working Group 2006). To meet this goal, it is necessary to assess and reduce the impact of analytical variability related to sample collection and processing on solubility estimates. Ideally, a standardized method for measuring aerosol TEI solubility would be adopted by the research community allowing for the straightforward intercomparison of analytical results. The work presented here was designed to examine the impact of filter substrate on measured fractional solubility. Eight filter substrates were compared to assess analytical variability introduced by filter composition. The TEI blank of each filter type was characterized to aid researchers in determining which filter substrate is best suited for their specific analytical needs. Aerosol solubility was estimated from measurements of the soluble and insoluble aerosol fractions. The soluble fractions of the major anions nitrate, sulfate, and oxalate were also measured. Sample handling was conducted following strict trace element clean protocols. The focus of this study was to determine the relative differences in aerosol solubility observations based on filter type and not to make determinations on the influence of environmental factors, both preceding and following deposition, on the respective solubilities of the elements and anions. Materials and procedures Materials ACS Plus grade HNO3 (70%) and HCl (38%) were purchased from Fisher Scientific and sub-boiling Teflon-distilled before use. HF (48%) was purchased in trace-metal grade. All references are to these acids at full concentration unless otherwise noted. All deionized water was ultrapure (pH ~5.5; ≥18.2 MW) produced from a NANOpure water system (Barnstead/Thermolyne). All calibration standards were made from SPEX multi-element and single element standards designed for mass spectrometry and were diluted in ultrapure water (UHP) to appropriate concentrations. All ion chromatograph (IC) standards were made from SPEX multi-anion standards and diluted in UHP to appropriate concentrations. Membrane filters were acid cleaned following two regimes. Polycarbonate and polytetrafluoroethylene filters were soaked for 1 week each in 4M HNO3, 3M HCl, and 0.5M sub-boiling distilled HCl. The filters were rinsed with UHP between each step. After the final acid washing, individual filters were rinsed again with UHP. The polyethersulfone and Metricel filters were acid washed following a similar protocol with the exclusion of the initial HNO3 wash. The GF/F and quartz filters were baked at 450° for 5 h, then placed in 3M HCl for 1 week. Following rinsing in UHP, the filters were soaked in 0.5M subboiling distilled HCl for an additional week followed by rinsing in UHP and drying. All drying was conducted in a Class 100 laminar flow hood. Blank filters of each type were processed in the same fashion as sample filters with the exception of field deployment. Aerosol collection Aerosol samples were collected using a total suspended particle (TSP) aerosol sampler located on the roof of the Interuniversity Institute of Marine Sciences in Eilat, Israel ~10 m off the northwest coast of the Gulf of Aqaba (29.5°N, 34.9°E). The aerosols were collected continuously over a 72-h period with an air flow of 1.2-1.5 m3 h–1. The sampler has four 47-mm filter cartridges connected to dedicated flow meters thereby collecting replicate samples simultaneously and allowing for the samples to be normalized by their respective volumes of filtered air. The four parallel samples are collected at slightly different flow rates but should be similar in composition because they are sampling the same air masses. The airflow path of the sampler and filter holders are made entirely of plastic to minBuck and Paytan Methods to assess aerosol TEI solubility