Method for determination of cesium-134 and -137 in seawater

Surface samples were pumped directly into new, calibrated 20 liter plastic cubitainers. Deeper samples (20-1000 m) were collected by CTD/Rosette. We found that the sample processing improved with 1 micromol prefiltration, presumably due to remove of organic particulates. This will not impact total Cs activities as the measured 137Cs and 134Cs on the 1 µm filters averaged only < 0.04% of the total during this cruise, consistent with the soluble nature of Cs in seawater.

Following collection, samples were acidified to pH 1-2 with JT Baker ultrapure nitric acid (130 mL per 22 L seawater). A stable 133Cs carrier (25 mg per mL standard, 1 mL per sample) was added for calibration of Cs extraction on the resin columns. Samples were allowed to equilibrate for at least 1 hour and an initial aliquot of sample was removed. Samples were extracted onto an ion exchange resin made of the organic polymer polyacrylonitrile, (PAN) and ammonium molybdophosphate (AMP) (1). Precleaned (with 0.1 N nitric acid) biorad glass chromatography columns (1.0x10 cm) were filled with 5 mL AMP-PAN resin in 0.1 N nitric acid. A frit was added above the resin to prevent re-suspension. Resin was allowed to settle for > 1 hour before processing. Samples were pumped directly from cubitainers through the resin column at a flow rate of 30-35 mL per minute. Eluted sample was collected into cleaned cubitainers and a final solution aliquot was taken to determine the column extraction efficiency by measuring stable 133Cs.

AMP-PAN resin was transferred to a vial and gamma counted while still wet for 134Cs and 137Cs isotopes using closed-end coaxial well detectors in the lab. Samples were counted until counting errors on all Cs peaks were generally below 5-10% (depending on sample activity). Samples with low activities were counted for 24-48 hours. Gamma spectrometers were calibrated with a range of 134Cs and 137Cs standards purchased from Eckert and Ziegler. The 137Cs peak at 661 keV and 134Cs peaks at 604 and 795 keV were identified and analyzed with Aptec software.

To check for column recoveries, initial and final sample aliquots containing stable cesium were diluted with 5% ultrapure nitric acid and analyzed on an Element 2 ICP-MS (Thermo Fisher Scientific, WHOI plasma facility) equipped with PFA MicroFlow nebulizer, quartz spray chamber and regular cones. Recovery corrections were applied to measured 134Cs and 137Cs activities with an average recovery of 93 plus/minus 5% standard deviation.

Calibration standards were run using this new AMP-PAN method for 137Cs. We found good agreement between archived water from the Sargasso Sea collected in 1978 (internal WHOI lab standard 137Cs = 3.4 ± 0.4 Bq m-3) and results of triplicate samples analyzed here (3.7 plus/minus 0.2 Bq m-3). Similarly, we measured a 137Cs activity in a single 5 L sample of 369 plus/minus 8 Bq m-3 in an IAEA Irish Sea water reference material, IAEA-443, which has a 95% confidence limit of 340-370 Bq m-3 (2).

The concentration of cesium isotopes in water is reported in units of Bq (1 Becquerel = one disintegration per second) per unit volume (Bq m-3). All activities are decay corrected to April 6th, the date of the maximum direct radioactivity discharge into the ocean (3).

Surface Cs measured by underway system is reported with the casts as the shallowest measurement and assigned a depth and pressure of 2.0.

BCO-DMO notes: Codes (parameter name WHOI_Cs_codes) were assigned to identify bottle sample type:

uw = underway Cs sample ns = not sample ind = Cs sample was drawn from a single bottle cb_X_X = Cs sample was drawn from multiple bottle identified by X and X