A new rapid protocol for 226Ra separation and preconcentration in natural water samples using molecular recognition technology for ICP-MS analysis
Graphical abstract
Introduction
Radium 226 is the most abundant radium isotope with a half-life of 1600 years present in uranium 238 decay chain. In the last few decades, uranium, coal or phosphate mining activities have released radium in the environment leading to radium activities significantly higher than the geochemical background level (Azouazi et al., 2001; Carvalho et al., 2007; Chalupnik and Wysocka, 2008; Cuvier et al., 2015; Pluta and Trembaczowski, 2001; Vandenhove et al., 2006). Due to its high specific activity and to its chemical properties favoring its accumulation in human bones increasing the internal radiation dose, 226Ra is considered to be one of the most radiotoxic naturally occurring isotopes (Jia and Jia, 2012; St-Amant et al., 2011). For this reason, it is important to understand its geochemical behavior and to identify the transfer pathways between the environment and humans. This requires being able to achieve high precision 226Ra measurements in environmental samples.
Reported 226Ra activities for natural water samples rarely exceed 30 mBq.L−1. Activities in river water generally range between 0.5 and 20 mBq.L−1 and for other surface areas, such as lakes are also within a narrow range (0.5 – 15 mBq.L−1), similar to that observed for river water (Vandenhove et al., 2019). Seawater 226Ra activities are particularly low ranging from 0.1 to 7 mBq.L−1 (Bourquin et al., 2011; IAEA, 2014) and groundwater activities are highly variable ranging from some tens of mBq.L−1 to several Bq.L−1 (IAEA, 2014).
Traditional methods for 226Ra analysis include radiometric techniques such as alpha and gamma spectrometry, radon 222 emanation and liquid scintillation (Godoy et al., 2016; Jia and Jia, 2012; Perrier et al., 2016; Song et al., 2017; van Beek et al., 2010). However, these techniques require large volume samples (≈ 100 mL to several liters), the achievement of secular equilibrium with radon 222 (≈ 30 days) and generally two to three days of counting for low 226Ra activities (Jia and Jia, 2012).
Initially, Inductively Coupled Plasma-Mass Spectrometry (ICP-MS) technique focused on longer-lived radionuclides (i.e. 238U, 232Th) where their low specific activities favored atom-counting over radiometric techniques. In the past few decades, improvements in instrument sensitivity achieved through advances in sample introduction, mass spectrometry configuration and vacuum pump technologies, have opened up the possibility of using ICP-MS for quantification of shorter-lived radionuclides, significantly reducing the analytical time for such analysis. During this period, 226Ra analysis with ICP-MS has grown significantly (Bourquin et al., 2011; Copia et al., 2015; Cozzella et al., 2011; Foster et al., 2004; Ghaleb et al., 2004; Hodge and Laing, 1994; Hsieh and Henderson, 2011; Joannon and Pin, 2001; Lagacé et al., 2017; Lariviere et al., 2003, 2005, 2007; Park et al., 1999; Pietruszka et al., 2002). However, one of the major drawbacks in analyzing 226Ra with this method is the presence of polyatomic interferences deteriorating the measurement accuracy (Epov et al., 2003; Lagacé et al., 2017; Lariviere et al., 2003, 2007; Park et al., 1999; van Es et al., 2017a, 2017b). Some of the most prominent ones according to the literature correspond to 88Sr138Ba+, 208Pb18O+,209Bi16O1H+, 40Ar40Ar146Nd+ and 40Ar186W+. 226Ra polyatomic interferences mainly originate from reactions occurring with the elements of the sample matrix in the plasma. In order to overcome this problem, an ICP-MS/MS with N2O as the reaction gas in the collision reaction cell has been successfully used in a recent work (Wærsted et al., 2018). However, chemical purification of samples before radium analysis remains the most commonly used procedure.
A wide variety of protocols and resins can be found in the literature capable of isolating and preconcentrating 226Ra, such as MnO2 resin (Moon et al., 2003), Mn-fiber (Bourquin et al., 2008), MnO2-PAN (Dulanská et al., 2015), TK100 resin (van Es et al., 2017b), Radium RAD Disks (Fons-Castells et al., 2017; Joannon and Pin, 2001), AnaLig® Sr-01 (Dulanská et al., 2016), water-soluble crown ethers (Chiarizia et al., 1999) and crown ether functionalized magnetic nanoparticles (Mesnic et al., 2013). Most of the protocols for ICP-MS applications use cation exchange resin such as AG-50W-X8 or Ln crown ether resin in combination with strontium specific resin such as Sr-Spec (Benkhedda et al., 2005; Chabaux et al., 1994; Copia et al., 2015; Cozzella et al., 2011; Joannon and Pin, 2001; Lariviere et al., 2005). To the best of our knowledge, no studies have considered the commercially available resin AnaLig® Ra-01 which was developed for Ra recognition. The aim of this work is therefore to test the resin with the objective to develop a rapid protocol for the separation and preconcentration of 226Ra in natural samples applicable to mass spectrometric measurements. Optimizations of 226Ra separation on AnaLig® Ra-01 and measurements by ICP-MS/MS were investigated in detail.
Section snippets
Reagents and materials
High purity acids were obtained by distillation (Savillex® DST-1000 system) from hydrochloric acid (HCl, Merck, Emsure 37%) and nitric acid (HNO3, VWR Chemicals, Normapur 68%). For render solution alkaline, ammonia (NH3, Suprapur 20%, Merck) was used. Deionized water was produced by a Millipore system (18.2 MΩ.cm resistivity). All sample dilutions and solutions for elemental analysis were performed with 0.5 mol.L−1 (M) HNO3. The same acid was used to prepare blanks and during the instrument
ICP-MS sensitivity performances
As can be seen in Fig. 2, the APEX-HF improved signal sensitivity for the Agilent 8800 compared to Scott spray chamber by a factor of 10. The LOD and LOQ were found to be 68 mBq.L−1 and 204 mBq.L−1 for the Scott spray chamber and 8 mBq.L−1 and 23 mBq.L−1 for the APEX, respectively. With the Scott spray chamber, sensitivity of radium was found to be similar to that obtained by van Es et al. (van Es et al., 2017b), around 9 CPS/(Bq.L−1). Results of LOD and LOQ obtained with ICAP-Q were close to
Conclusion
This work presents a new rapid 226Ra separation protocol using the Analig® Ra-01 resin. Batch and column results showed high specificity for 226Ra in a wide range of acid concentrations to 0.01 M from 10 M. Radium competitor elements were found to be only Sr, Ba, Tl and Pb. The determination of distribution coefficients allowed identifying conditions for their selective elution from radium. Radium elution was performed with 0.12 M NTA. Ba coelutes with Ra, however Ba itself cannot interfere
Acknowledgements
We would like to warmly thank Nicolas Cariou for his expertise in Element 2 ICP-MS, Azza Habibi and Olivier Diez for many helpful discussions on chromatographic conditions for element separation, Sylvain Bassot and the LEI, IRSN laboratory, for their helpful contribution for 226Ra standard solution acquisition in our laboratory. We would like also to warmly thank Charlotte Cazala, head of the LEI laboratory, for her support for this project. This is PATERSON, the IRSN's mass spectrometry
References (41)
- et al.
Natural radioactivity in phosphates, phosphogypsum and natural waters in Morocco
J. Environ. Radioact.
(2001) - et al.
Determination of 226Ra concentrations in seawater and suspended particles (NW Pacific) using MC-ICP-MS
Mar. Chem.
(2011) - et al.
Comparison of techniques for pre-concentrating radium from seawater
Mar. Chem.
(2008) - et al.
Radioactivity in the environment around past radium and uranium mining sites of Portugal
J. Environ. Radioact.
(2007) - et al.
A new Ra-Ba chromatographic separation and its application to Ra mass-spectrometric measurement in volcanic rocks
Chem. Geol.
(1994) - et al.
Radium removal from mine waters in underground treatment installations
J. Environ. Radioact.
(2008) - et al.
Determination of 226Ra in urine samples by Q-ICP-MS: a method for routine analyses
Radiat. Meas.
(2011) - et al.
Uranium decay daughters from isolated mines: accumulation and sources
J. Environ. Radioact.
(2015) - et al.
Simultaneous determination of 226Ra, 228Ra and 210Pb in drinking water using 3M Empore™ RAD disk by LSC-PLS
Appl. Radiat. Isot.
(2017) - et al.
226 Ra and Ba concentrations in the Ross Sea measured with multicollector ICP mass spectrometry
Mar. Chem.
(2004)