Dataset: GA01 - Dissolved Pb
Deployment: GEOVIDE

Sample storage bottle lids and threads were soaked overnight in 2N reagent grade HCl, then filled with 1N reagent grade HCl to be heated in an oven at 60 degrees Celsius overnight, inverted, heated for a second day, and rinsed 5X with pure distilled water.  The bottles were then filled with trace metal clean dilute HCl (0.01N HCl) and again heated in the oven for one day on either end.  Clean sample bottles were emptied, and double-bagged prior to rinsing and filling with sample.

As stated in the cruise report, trace metal clean seawater samples were collected using the French GEOTRACES clean rosette (General Oceanics Inc. Model 1018 Intelligent Rosette), equipped with twenty-two new 12L GO-FLO bottles (two bottles were leaking and were never deployed during the cruise). The 22 new GO-FLO bottles were initially cleaned in LEMAR laboratory following the GEOTRACES procedures (Cutter and Bruland, 2012). The rosette was deployed on a 6mm Kevlar cable with a dedicated custom designed clean winch. Immediately after recovery, GO-FLO bottles were individually covered at each end with plastic bags to minimize contamination. They were then transferred into a clean container (class-100) for sampling. On each trace metal cast, nutrient and/or salinity samples were taken to check potential leakage of the Go-Flo bottles. Prior to filtration, GO-FLO bottles were mixed manually three times. GO-FLO bottles were pressurized to less than 8 psi with 0.2-um filtered N2 (Air Liquide). For Stations 1, 11, 15, 17, 19, 21, 25, 26, 29, 32 GO-FLO spigots were fitted with an acid-cleaned piece of Bev-a-Line tubing that fed into a 0.2 um capsule filters (SARTOBRAN 300, Sartorius). For all other stations (13, 34, 36, 38, 40, 42, 44, 49, 60, 64, 68, 69, 71, 77) seawater was filtered directly through paired filters (Pall Gelman Supor 0.45um polystersulfone, and Millipore mixed ester cellulose MF 5 um) mounted in Swinnex polypropylene filter holders, following the Planquette and Sherrell (2012) method. Filters were cleaned following the protocol described in Planquette and Sherrell (2012) and kept in acid-cleaned 1L LDPE bottles (Nalgene) filled with ultrapure water (Milli-Q, 18.2 megaohm/cm) until use. Subsamples were taken into acid-cleaned (see above) Nalgene HDPE bottles after a triple rinse with the sample. All samples were acidified back in the Boyle laboratory at 2mL per liter seawater (pH 2) with trace metal clean 6N HCl.

On this cruise, only the particulate samples were assigned GEOTRACES numbers. In this dataset, the dissolved Pb samples collected at the same depth (sometimes on a different cast) as the particulate samples have been assigned identifiers as “SAMPNO” which corresponds to the particulate GEOTRACES number. In cases where there were no corresponding particulate samples, a number was generated as “PI_SAMPNO”.

Upon examining the data, we observed that the sample taken from rosette position 1 (usually the near-bottom sample) was always higher in [Pb] than the sample taken immediately above that, and that the excess decreased as the cruise proceeded. The Pb isotope ratio of these samples were higher than the comparison bottles as well. A similar situation was seen for the sample taken from rosette positions 5, 20 and 21 when compared to the depth-interpolated [Pb] from the samples immediately above and below. Also, at two stations where our near-bottom sample was taken from rosette position 2, there was no [Pb] excess over the samples immediately above. We believe that this evidence points to sampler-induced contamination that was being slowly washed out during the cruise, but never completely. So we have flagged all of these analyses with a “3” indicating that we do not believe that these samples should be trusted as reflecting the true ocean [Pb].

In addition, we observed high [Pb] in the samples at Station 1 and very scattered Pb isotope ratios. The majority of these concentrations were far in excess of those values observed at nearby Station 11, and also the nearby USGT10-01. Discussion among other cruise participants revealed similarly anomalous data for other trace metals (e.g., Hg species). After discussion at the 2016 GEOVIDE Workshop, we came to the conclusion that this is*- evidence of GoFlo bottles not having sufficient time to “clean up” prior to use, and that most or all bottles from Station 1 were contaminated. We flagged all Station 1 data with a “3” indicating that we do not believe these values reflect the true ocean [Pb].

Samples were analyzed at least 1 month after acidification over 36 analytical sessions by a resin pre-concentration method. This method utilized the isotope-dilution ICP-MS method described in Lee et al. 2011, which includes pre-concentration on nitrilotriacetate (NTA) resin and analysis on a Fisons PQ2+ using a 400uL/min nebulizer. Briefly, samples were poured into 30mL subsample bottles. Then, triplicate 1.5mL polypropylene vials (Nalgene) were rinsed three times with the 30mL subsample.  Each sample was pipetted (1.3mL) from the 30mL subsample to the 1.5mL vial.  Pipettes were calibrated daily to the desired volume.  25 ul of a 204Pb spike were added to each sample, and the pH was raised to 5.3 using a trace metal clean ammonium acetate buffer, prepared at a pH of between 7.95 and 7.98.  2400 beads of NTA Superflow resin (Qiagen Inc., Valencia, CA) were added to the mixture, and the vials were set to shake on a shaker for 3 – 6 days to allow the sample to equilibrate with the resin.  After equilibration, the beads were centrifuged and washed 3 times with pure distilled water, using a trace metal clean siphon tip to remove the water wash from the sample vial following centrifugation.  After the last wash, 350μl of a 0.1N solution of trace metal clean HNO3 was added to the resin to elute the metals, and the samples were set to shake on a shaker for 1 – 2 days prior to analysis by ICP-MS.

NTA Superflow resin was cleaned by batch rinsing with 0.1N trace metal clean HCl for a few hours, followed by multiple washes until the pH of the solution was above 4.  Resin was stored at 4 degrees Celsius in the dark until use, though it was allowed to equilibrate to room temperature prior to the addition to the sample.

Nalgene polypropylene (PPCO) vials were cleaned by heated submersion for 2 days at 60 degrees Celsius in 1N reagent grade HCl, followed by a bulk rinse and 4X individual rinse of each vial with pure distilled water. Each vial was then filled with trace metal clean dilute HCl (0.01N HCl) and heated in the oven at 60 degrees Celsius for one day on either end.  Vials were kept filled until just before usage.

On each day of sample analysis, procedure blanks were determined. Replicates (12) of 300uL of an in-house standard reference material seawater (low Pb surface water) were used, where the amount of Pb in the 300uL was verified as negligible. The procedural blank over the relevant sessions for resin preconcentration method ranged from 2.2 – 9.9pmol/kg, averaging 4.6 +/- 1.7pmol/kg. Within a day, procedure blanks were very reproducible with an average standard deviation of 0.7pmol/kg, resulting in detection limits (3x this standard deviation) of 2.1pmol/kg. Replicate analyses of three different large-volume seawater samples (one with 11pmol/kg, another with 24pmol/kg, and a third with 38pmol/kg) indicated that the precision of the analysis is 4% or 1.6pmol/kg, whichever is larger.

Triplicate analyses of an international reference standard gave SAFe D2: 27.2 +/- 1.7 pmol/kg. However, this standard run was linked into our own long-term quality control standards that are run on every analytical day to maintain long-term consistency.  

For the most part, the reported numbers are simply as calculated from the isotope dilution equation on the day of the analysis. For some analytical days, however, quality control samples indicated offsets in the blank used to correct the samples. For the upper 5 depths of Station 29, all depths of Station 40, and the deepest 2 depths of Station 42, the quality control samples indicated our blank was overcorrecting by 3.4pM, and we applied a -3.4pM correction to our Pb concentrations for that day. For the deepest 11 depths of Station 34, the quality control samples indicated our blank was overcorrecting by 10.2pM (due to contamination of the low trace metal seawater stock), and we applied a -10.2 pM correction to our Pb concentrations for that day. With these corrections, the overall internal comparability of the Pb collection should be better than 4%.

The errors associated with these Pb concentration measurements are on average 3.2% of the concentration (0.1 – 4.4pmol/kg). Although there was a formal crossover station (1) that overlaps with USGT10-01 (GA-03), sample quality on the first station of GEOVIDE appears problematical making the comparison unhelpful. However, GEOVIDE station 11 (40.33 degrees North, 12.22 degrees West) is not too far from USGT10-01 (38.325 degrees North, 9.66 degrees West) and makes for a reasonable comparison. It should also be noted that the MIT lab has intercalibrated Pb with other labs on the 2008 IC1 cruise, on the 2011 USGT11 (GA-03) cruise, and on the EPZT (GP-16) cruises, and maintains in-lab quality control standards for long-term data quality evaluation.

Ten percent of the samples were analyzed by Rick Kayser and the remaining ninety percent of the samples were analyzed by Cheryl Zurbrick. There was no significant difference between them for the lowest concentration large-volume seawater reference sample (RK averaged 11.0 +/- 1.6 pmol/kg; CZ averaged 11.6 +/- 1.6 pmol/kg).