Comparison of methods to determine extraction efficiencies of Ra isotopes and 227Ac from large volume seawater samples

Radium isotopes, other than 226Ra, and 227Ac are typically present at low activities in the open ocean. The analysis of these isotopes thus requires collection of large volumes of seawater and high sensitivity, low background instruments. To obtain the required large volumes (hundreds to thousands of liters), these radionuclides are typically preconcentrated on cartridge-style filters impregnated with MnO2 (Mn-cartridges) deployed on in-situ pumps. This technique, however, requires the determination of the extraction efficiency of the Mn-cartridges for the radionuclides of interest. For Ra isotopes, we used two methods to estimate the extraction efficiency of these Mn-cartridges at two stations on the South-West Indian Ridge in the Southern Ocean (GEOTRACES GS02). Method (1) compares the 226Ra activities recovered on the Mn-cartridges versus the activities determined in Mn-fibers, through which seawater was passed at a flow rate < 1 L min−1 to quantitatively sorb Ra (Mn-fiber method) while method (2) combines the 226Ra activities determined from two Mn-cartridges placed in series on in-situ pumps (A-B method). The second method is also applied to determine the 227Ac extraction efficiency. We find a relatively wide-range of Ra and 227Ac extraction efficiencies across the dataset (from 44.8% to 99.6% for Ra, and from 23.7% to 77.5% for 227Ac). Overall, the yield of 227Ac extraction is lower than that of Ra (mean value of 49.3 ± 19.0% for 227Ac, n = 10; mean value of 79.2 ± 10.3% for Ra, n = 13, using the Mn-fiber method and a mean value of 63.9 ± 12.5%, n = 11 using the A-B method). Our dataset suggests that the Ra extraction efficiencies using either the A-B method or the Mn-fiber method are in relatively good agreement. Consequently, the 223Raex, 224Raex and 228Ra activities determined from the Mn-cartridges by applying the two Ra extraction yields are similar. We also show that the 227Ac extraction efficiency can be estimated from the Ra extraction efficiency allowing the use of a single Mn-cartridge. Finally, we recommend to determine the Ra and 227Ac extraction efficiencies in each individual Mn-cartridge, rather than applying a single extraction efficiency to all the Mn-cartridges, since a significant variability in the extraction efficiencies was observed between the different Mn-cartridges.