Effects of flow rates and composition of the filter, and decay/ingrowth correction factors involved with the determination of in situ particulate 210Po and 210Pb in seawater

Accurate measurements of particulate 210Po (Pop) and 210Pb (Pbp) are required in the investigation of (i) partitioning of Po and Pb between particulate and dissolved phases and (ii) export estimates of carbon and other key trace metals from the euphotic zone via sinking particulate matter. Based on the intercomparison of different composition of the filter material (QMA, Supor, Millipore, and Pall GN6) and flow rates (2 to 8 L min), we show how these factors affect the measured concentrations of Pop and Pbp and their activity ratios (AR). As such, we recommend using Supor 0.4 μm filter and a flow rate of up to 8 L/min for the measurements of Pop and Pbp. Furthermore, we inter-compared Pop and Pbp obtained by small-volume McLane and large-volume MULVFS pumps. The activities of Pop in MULVFS filter samples are somewhat higher than that of McLane filter samples, whereas the 210Po/210Pb AR collected by McLane pump is distinctly lower, suggesting some fractionation in the collection process by the pumping systems. Likewise comparison of vertical profiles of Pop and Pbp obtained using McLane pumps by two independent research groups at the two intercalibration stations in the Pacific Ocean show quite similar values except in the mesopelagic waters, suggesting possible uneven loading and presence of larger gelatinous plankton in the filter. Finally, we append a detailed analysis of various correction factors for the accurate calculation of in situ 210Po and 210Pb. Presented results are relevant to the worldwide community that uses 210Po-210Pb as tracers in aquatic systems. *Corresponding author: E-mail: [email protected] Acknowledgments We thank a large number of people who assisted with the sample collection. We thank Jim Bishop for providing an aliquot of filter samples from MUVLFS pumping system. This work was supported by NSF (OCE0851032, MB), NSF (OCE-0851462, TC). Gi-Hong’s sabbatical leave at Wayne State University was supported by Korea Ocean Research and Development Institute. Ma Qiang was supported by a 1 year Scholarship by Marine Pubic Welfare Project (201005012-3) and Fujian Natural Science Foundation (2009J06026) to visit WSU. This paper is a part of the Intercalibration in a special issue of Limnology and Oceanography: Methods, and the special issue is supported by funding from the U.S. National Science Foundation, Chemical Oceanography Program. DOI 10.4319/lom.2013.11.126 Limnol. Oceanogr.: Methods 11, 2013, 126–138 © 2013, by the American Society of Limnology and Oceanography, Inc. LIMNOLOGY and OCEANOGRAPHY: METHODS Baskaran et al. Particulate 210Po and 210Pb measurement 127 pore-size filter. The dissolved and particulate phases are defined operationally by the pore size of the filter used. The most widely used pore sizes range from 0.2 to 1.0 μm. A majority of the filters also have only nominal pore sizes rather than absolute cut-off. The pore size, filtration rate (which is a measure of the contact time of the water containing dissolved species with the filter and also a measure of the pressure exerted onto particles during filtration) and the chemical composition of the filter (due to possible sorption of radionuclide onto filter matrix) could affect the measured concentrations of particulate radionuclides and may differ from the in situ concentration. To assess the effects of pore size, flow-rates, and the composition of the filter material on the measured concentrations of particulate 210Po (Pop) and 210Pb (Pbp), we conducted a series of experiments with different pore sizes, flow rates, and filter composition to answer the following questions: (1) How do the concentrations of Pop and Pbp vary with different pore sizes (0.2, 0.4 and 0.8 μm Supor), flow rates (2.25 to 8.25 L/min through 1 μm QMA) and composition of the filter (QMA, Supor, Millipore HA, and Pall GN6)? (2) How do the Pop and Pbp concentrations compare in samples collected using small-volume pumps (4-8 L/min) with that in the large-volume pumps (MULVFS, 59 L/min)? (3) What is the distribution between the concentrations of Pop and Pbp measured in the top filter (F1) to that in the back-up filter (F2) and what are the possible sources of these artifacts; and (4) How do the concentrations of Pop and Pbp in vertical profiles from the Pacific inter-calibrated between two laboratories compare? In a related paper (Church et al. 2012), we presented analytical results from the intercalibration of dissolved and particulate samples at several GEOTRACES intercalibration sites in the North Atlantic and Pacific. In this article, we append for general use the various factors and assumptions in calculating final 210Po and 210Pb data, such as correction factors due to in situ decay and growth of final 210Po from the presence of 210Pb before and between times of separation. Materials and procedures General considerations The chemical procedure and radiochemical assay followed by most of the research groups for the measurement of 210Po and 210Pb in seawater (either particulate and/or dissolved) are mostly similar (Fleer and Bacon 1984; Friedrich and Rutgers van der Loeff 2002; Hong et al. 1999; IAEA 2009; Kim et al. 1999; Masque et al. 2002; Mathews et al. 2007; Radakovitch et al. 1998; Rutgers van der Loeff and Moore 1999; Sarin et al. 1992; Stewart et al. 2007). The determination of 210Po and 210Pb in seawater (either particulate or dissolved phase) is routinely conducted in the same sample, first by measuring 210Po (called ‘in situ’ 210Po) and then keeping the same sample for a period of about 6 months to 2 years for the ingrowth of 210Po from the decay of in situ 210Pb. The second 210Po (called ‘parent-supported’) measurement provides the data on 210Pb. The most widely used instrument for seawater (both dissolved and particulate) 210Po and 210Pb analysis is an a spectrometer with surface barrier detectors (see reviews in Mathews et al. 2007 and Baskaran et al. 2009a). Currently, the recommended assay of both dissolved and particulate sea water samples are outlined in the GEOTRACES procedures manual (GEOTRACES 2010) Sampling of particulate matter Details on the collection of small-volume (McLane) and large-volume (MULVFS) particulate matter as well as sub-sampling of the filters are given in Maiti et al. (2012) and Bishop et al. (2012). Water samples were filtered through McLane in situ pumps from two stations during IC-2 cruise: North Pacific at SAFe (Sampling and Analysis of Iron) (30°00¢N; 140°00¢W, 4000 m depth) and Santa Barbara Basin (SBB; 34°17¢N; 120°03¢W, 850 m depth) stations. All the filters (QMA-Quartz microfiber membrane, Supor-polyethersulfone membrane, Millipore HA-mixed cellulose ester membrane and Pall GN6mixed nitrocellulose membrane) used in the present study are nominal pore sizes and were acid leached with 10% HCl, to remove any metals present in the filter, in a shore-based laboratory before use in the field. The McLane pumps were programmed to operate within a range of 4-8 L/min. The QMA filters used in MULVFS were also acid-leached prior to use in the field (Bishop et al. 2012). For Pop and Pbp, filtering through a conventional filtration set-up, such as passing the requisite volume (10s L) through 0.45 μm Supor membrane filters, is time consuming. Also longer contact time of the seawater with the filter material due to filter loading or clogging could result in the sorption of dissolved and colloidal 210Po and/or 210Pb onto the filter. Although capsule filters are more efficient with respect to filtration, quantitative retrieval of particulate matter from such filter cartridges is quite difficult. Results from the IC-1 cruise for Pop and Pbp indicate that 10-20 L water samples from open ocean filtered through 380 mm2 area of the filter paper (22-mm diameter circular punch from a 142-mm diameter filter) have a relatively high error on the activities of 210Po and 210Pb (>20%; Church et al. 2012). Digestion of the filter sample In the open ocean, the suspended particulate matter is usually comprised of < 5% lithogenic material (Chester 1990) and hence most of the Pb and Po are present in association with biogenic particulate matter. In the case of 210Po, there could be some possible transport within the cell material in the biota (Stewart and Fisher 2003a, 2003b). A number of procedures have been followed in the digestion of the filter material. Since the 210Poand 210Pb-laden particulate matter is retained on the filter material, digestion with a combination of HF (to break the Si matrix), HNO3 (to break the organic matrix), and HCl (to convert to chloride medium) should be sufficient. It is likely that almost all of the biogenic matter is decomposed by concentrated nitric acid digestion method. The Quartz Microfiber (QMAs, with a nominal pore size) as well as glass Baskaran et al. Particulate 210Po and 210Pb measurement 128 fiber filter (GFFs, with a typical nominal pore size) were found to be totally dissolved with HF-HNO3-HCl acid digestion while Supor filters resulted in some amount of residual material. Because most of the particulate matter is biogenic, we do not recommend use of dissolution with HClO4, which requires a special fume-hood that may not be readily available with many research groups. A flow chart showing the analysis of filters for Pop and Pbp is given in Fig. 1. The filter containing the particulate matter was transferred to a 100 mL Teflon beaker, and then a known amount of 209Po (~1 dpm) SRM Standard (NIST-SRM4326) and 1 mg stable Pb (AAS Standard) were added as yield monitor (Pb yield monitor for the entire chemical processing steps including ion-exchange separation of Pb and Po). One of the 8 intercalibration groups used 208Po spike, although the a energy difference between 209Po and 210Po is higher (423.5 keV) than that of 208Po and 210Po (189.4 keV) allowing better resolution with 209Po spike. Furthermore, it is widely recognized that 208Po spike contains some amount of 209Po impurity, and hence those groups that use double spike method (209Po for in situ 210Po and 208Po for in situ 210Pb via 210Po determination) will have to take the presence of 209Po in the 208Po spike in to consideration, because the correction for 209Po present in 208Po spike becomes more serious with time as the decay rate of 208Po [t1/2 = 2.898 years] is more than 35 times faster than that of 209Po [t1/2 = 102.5 years]). Two digestion methods were employed: (i) In the open digestion method, 5 mL each of trace-metal grade HF, conc. HNO3 and conc. HCl was added to the Teflon beaker after adding a known amount of 209Po (~1 dpm) and stable Pb (1 mg) and the solution was digested at ~90°C on a hot plate. After the solution completely dried, 5 mL each of HF-HNO3Fig. 1. Flow chart for the analysis of particulate 210Po and 210Pb. Baskaran et al. Particulate 210Po and 210Pb measurement 129 HCl digestion was repeated twice and to the final residue, another 5 mL of conc. HCl was added and dried; the residue was finally taken in 5 mL 6M HCl; and (ii) in the closed digestion method, 10 mL each of conc. HF-HNO3HCl were added to the Teflon digestion vessel after adding a known amount of 209Po (~1 dpm) and 1 mL of stable Pb (= 1 mg Pb). The solution was digested at 100°C for 6 h; the digested solution was dried followed by another drying with 5 mL conc. HCl. The final residue obtained from both open and closed digestion methods was taken in 5 mL 6M HCl. To this solution, 20 mL of distilled water was added and the solution was centrifuged to remove solid residue, if needed. This 25 mL of 1.2 M HCl solution was used for electroplating on to a Ag planchet for 4 h at 90°C. A detailed discussion on optimizing various factors to obtain maximum plating efficiency is given in Lee et al. (in press). The GEOTRACES procedures manual (GEOTRACES 2010) recommends plating each sample for at least 4-6 h at 80-90°C, although a wide range of temperatures (80°C-90°C) and time (1.5-24 h) have been reported (summarized in Mathews et al. 2007; Lee et al. in press). Quantitative separation of 210Po and 210Pb after plating for the assay of in situ 210Pb To determine the in situ 210Pb, the plated solution used for in situ 210Po is stored for 6 to 24 months for the ingrowth of 210Po and the ingrown 210Po is again assayed by most laboratories. After the plating of in situ 210Po, often some amount of Po (210Po and the spike, 209Po) is left behind in the solution. Many research groups use additional Ag plates to remove the residual Po, but this is reported to remove only a portion of the residual 210Po (Cochran et al. 1983). Anion exchange resin column separation will only ensure complete separation of Po from Pb (Chung et al. 1983; GEOTRACES procedure manual [GEOTRACES 2010]). Some research groups assume that allowing the plated solution to sit for 1-2 y will result in the decay of 210Po to negligible amounts (in 1-2 y, 84% to 94% of the residual 210Po would have decayed), but there is very little decay of 209Po (0.7% to 1.3%) that takes place during this period. Therefore, when 209Po spike is also used for the determination of in situ 210Pb via 210Po, the presence of residual 209Po will underestimate the results. The effect of residual Po on the determination of activity of in situ 210Pb is graphically shown in Fig. 2 under different scenarios based on varying amounts of residual Po and time for the ingrowth of 210Po from in situ 210Pb, assuming that the activity ratio 210Po/210Pb = 1 in the sample; spike 209Po activity = in situ 210Po activity (Fig. 2). The percentage difference in the 210Po/209Po ratios with and without residual Po is plotted as a function of time (Fig. 2). It is evident that quantitative removal of Po from the plated solution, which is used for in situ 210Pb determination, is required. The chemical procedure (Fig. 1) that we have adopted (recommended in the GEOTRACES procedure manual [GEOTRACES 2010]) involves ion-exchange column separation (DOWEX-1, anion exchange AG-1-X8 [100-200 mesh], 9M HCl column, e.g., Baskaran et al. 2009a) of Po and Pb in the presence of HCl to effect passage of the 210Pb as the chlorocomplex and the quantitative retention of 210Po and 209Po on to resin beads.