Low-level 226Ra determination in groundwater by SF-ICP-MS: optimization of separation and pre-concentration methods
© Copia et al.; 2015
Received: 22 May 2015
Accepted: 2 June 2015
Published: 25 June 2015
Inductively coupled plasma mass spectrometry (ICP-MS) techniques have been widely used for analysis of long-lived environmental radionuclides. In this paper, we present an optimization of the sector field (SF)-ICP-MS technique for the analysis of 226Ra in groundwater samples using a method of pre-concentration of radium in water samples.
The separation protocol and a sequential application of ion exchange and extraction chromatography have been optimized, and related polyatomic interferences and matrix effects affecting the 226Ra signal were investigated.
Analyzing 12 replicates (water spiking at 22 fg g−1 of 226Ra), the 226Ra recovery efficiency close to 100 % has been obtained. The instrumental 226Ra detection limit of 0.09 fg g−1 (3σ criterion) and the absolute detection limit of 0.05 fg in a 25-mL groundwater sample have been reached.
An optimization of the radium separation method and a pre-concentration of radium in groundwater samples led to high radium recoveries, almost up to 100 %. The same could be said with respect to the separation of the interfering elements, important for the quantitative 226Ra analysis by SF-ICP-MS. The improvements in the separation and pre-concentration techniques also helped to improve the 226Ra detection limit down to 0.05 fg/25 mL of groundwater sample.
The radium-226, 226Ra (T ½ = 1622 years), is a naturally occurring radioisotope belonging to the 238U decay chain, which has widely been used as a tracer in groundwater and seawater transport and mixing (Smith et al. 2012), as well as for studying water-rock interactions in groundwater reservoirs (Reynolds et al. 2003). Several analytical radiometric and mass spectrometry techniques have been developed for the analysis of 226Ra in environmental samples: radon emanation (Kim et al. 2001), liquid scintillation spectrometry (Repinc and Benedik 2002), alpha-ray spectrometry (Morvan et al. 2001), gamma-ray spectrometry (Johnston and Martin 1997), thermal ionization mass spectrometry (TIMS) (Cohen and O’Nions 1991), and inductively coupled plasma mass spectrometry (ICP-MS) (Chabaux et al. 1994; Park et al. 1999; Lariviere et al. 2003; Lariviere et al. 2005; Zheng and Yamada 2006; Varga 2008; Cozzella et al. 2011; Hou and Roos 2008; Butler et al. 2015). With the exception of ICP-MS, these methods are time consuming, often expensive, and therefore not convenient when large number of samples are to be analyzed. However, the ICP-MS method has also some drawbacks, first of all possible interferences, which could make precise ICP-MS analysis of long-lived radionuclides in the environment difficult (Wyse et al. 2001; Lee et al. 2008; Lehto and Hou 2010; Lujanienė et al. 2013).
The aim of the present work has been to develop a robust technique for the 226Ra analysis of groundwater samples collected from the Gran Sasso (GS) aquifer (central Italy) (Plastino et al. 2013). Within the framework of the Environmental Radioactivity Monitoring for Earth Sciences (ERMES) project carried out at the Instituto Nazionale di Fisica Nucleare (INFN) Gran Sasso National Laboratory, 1-L groundwater samples have been collected weekly since 2008 at ten different sites located in the underground laboratory (Plastino et al. 2010; Plastino et al. 2011). In order to analyze with high precision such large number of samples necessary for the development of 226Ra time series, it has been proposed to develop a new analytical technology, which would be based on ICP-MS.
To accomplish the above requirements, a method described by Lariviere et al. (2005) has been chosen as a suitable base for further developments. This method consists of the sequential application of a cation exchange resin (50W-X8) and an extraction chromatographic resin (Sr resin). The first method has been widely used to separate 226Ra in liquid matrices (Lariviere et al. 2005; Varga 2008; Cozzella et al. 2011; Hou and Roos 2008; Butler et al. 2015), while the latter retains Sr and partially also Ba, the main polyatomic interferences in the analysis of 226Ra in groundwater samples (Chabaux et al. 1994; Horwitz et al. 1992). Due to high-precision data requirements for the development of the 226Ra groundwater time series, the original Lariviere et al. (2005) method has been modified by adding a pre-concentration part, effects of which were thoroughly investigated. This work focuses therefore on the optimization of the separation protocol, on the pre-concentration of samples, and on the evaluation of particularly related polyatomic interferences and matrix effects that could affect the 226Ra peak window.
Reagents and standards
Deionized water (Milli-Q water system, Millipore, Bedford, USA) and hyperpure hydrochloric and nitric acids (Panreac, Barcelona, Spain) were used during this study. Mono-elemental standard of Ca, Ba, Mg, Sr, Ce, La, and Nd (NIST, Gaithersburg, USA) in stock solutions were diluted to obtain matrix-matched standards and spikes. Radium standard and spikes were obtained by dilution from a 100.5 Bq g−1 STD (PTB, Braunschweig, Germany). Pre-concentration and separation of 226Ra were performed using two commercially available resins: a strong cation exchange resin (AG-50W-X8) and a crown ether-based resin (Sr resin) (Triskem, Bruz, France).
Optimized values of Finnigan Element 2 with Apex Q introduction system
Gas flow (L min−1)
RF power (W)
Number of run
Number of passes
Samples per peak
Mass window (%)
Integration window (%)
Sample time (s)
Acquisition time (s)
Apex Q parameters
Spray chamber temperature (°C)
Peltier-cell condenser temperature (°C)
The optimization of the procedures for the ICP-MS analysis of 226Ra in groundwater samples was carried out with the aim to accomplish separation of radium from matrix constituents and from spectral interfering elements. One gram of (AG-50W-X8) resin was packed into a plastic cartridge. The column was pre-washed and conditioned using 10 mL of 4-M HNO3, 5 mL of H2O Milli-Q, and 10 mL of 1.7-M HCl. Then, 25 mL of acidified (pH 1 HCl) sample was loaded. After that, 15 mL of 2.5-M HCl were eluted to wash away Ca, Mg, and the other matrix constituents.
At this point, the AG-50W-X8 and Sr resin cartridges were connected in series, and 10 mL of 4-M HNO3 were eluted through to desorb Ra from AG-50W-X8 and separate it from Sr and Ba in the Sr resin. Then, 10 mL of 3-M HNO3 were eluted in order to completely recover Ra. At the end of this procedure, the solution containing Ra was collected, evaporated, and re-dissolved in 0.5 mL of 3-M HNO3.
Results and discussion
Starting from the a priori assumption that 238U decay chain is in secular equilibrium in the investigated groundwater samples, and having a 238U concentration between 1 and 2 ng g−1, a 226Ra concentration around 0.3–0.7 fg g−1 was expected. At such a low concentration, radium was barely detectable by available instrumentation, so a pre-concentration was needed. A pre-concentration factor of 50 was chosen because, on one side, it guarantees a measurable signal and a good counting statistics; on the other hand, a small sample amount of 25 mL of water is sufficient for the analysis. The 226Ra ICP-MS measurements suffer of both spectral and nonspectral interferences; therefore, it is crucial to understand how these elements are separated in the procedure, how big are their levels in the final solution, and which are their effects on the measurement.
Polyatomic interferences studies
The blank (BLK) and Gran Sasso (GS) groundwater signal values (columns 2 and 3)
GS sample (s−1)
Sr resin (s−1)
149 · 103
97 · 106
3 · 106
120 · 106
44 · 103
51 · 106
490 · 103
19.7 · 106
63.9 · 106
1.8 · 103
90.5 · 103
960 · 103
41.5 · 103
910 · 103
1.4 · 103
68 · 103
200 · 103
Possible interferences from Ce, La, and Nd were evaluated measuring a solution containing 50 pg g−1 of Ce, La, and Nd and 10 ng g−1 of Sr. These concentrations were around ten times higher than actual values in the final step of the procedure. As shown in Table 2, no sign of interferences were observed.
Typical Ca, Mg, K, and Na concentration values in a sample of Gran Sasso (GS) groundwater (column 2), theoretical values after pre-concentration without separation (column 3), Ca and Mg signals in GS groundwater (column 4), after elution procedure (column 5), after evaporation (column 7), and the relevant separation efficiencies (column 6)
GS (μg g−1)
GSAPCWS(x) (μg g−1)
After proc. (s−1)
Sep. eff. (%)
After evap. (s−1)
46 · 106
0.26 · 106
10.4 · 106
860 · 106
1.63 · 106
65.2 · 106
Ca (μg g−1)
Mg (μg g−1)
These effects preclude a priori the application of chromatographic extraction methods without the separation of matrix elements, as for example in the Lariviere et al. (2003) method, or in the mere application of a Sr resin in order to eliminate polyatomic interferences. After application of the elution procedure, the matrix elements were separated with efficiency higher than 99 %. Then, the sample was concentrated by a factor 40, obtained evaporating to dryness the 20 mL of elution products and re-dissolving in 0.5 mL of 3-M HNO3. Signal values and relevant efficiencies are reported in Table 3. The matrix effect was then evaluated for concentrations around those previously obtained, using five matrix-matched 226Ra standards, with increasing concentrations of Ca (1, 2, 5, 10, and 20 μg g−1) and Mg (0.25, 0.5, 1.25, 2.5, and 5 μg g−1), and comparing them with a standard without Ca and Mg, at the same 226Ra concentration (27.5 fg g−1). The obtained results are also reported in Table 3. Results showed that the presence of the matrix has a positive effect on the signal, remaining quite constant over the concentration range. This can be taken into account in measurement sessions performing a calibration curve acquisition with standards matching the Ca and Mg proposed concentrations.
Optimization of separation procedures and recovery studies
pre-washing and conditioning
washing away Ca, Mg, and other matrix constituents using HCl
recovering of Ra using HNO3
The Ra recovery efficiencies and the Ba, Sr, Ca, and Mg separation efficiencies
Recovery eff. (%)
Separation efficiency (%)
After washing with HCl (step 3), the two cartridges were connected in series and the rest of the procedure was performed. The series connection allows to avoid an intermediate evaporation step in the Lariviere et al. (2005) procedure, with a significant reduction of its duration. Loading a sample in the Sr resin at 4-M HNO3 did not affect the column capacity factor k’, as previously described (Plastino et al. 2010). After the elution in Sr resin, as shown in Fig. 3, the separation efficiencies of Ba and Sr were 98 and 99.3 %, respectively, and the resulting values of Ba and Sr were below the level of interference formation (Fig. 1). The amount of 3-M nitric acid rinse was then reduced to 5 mL.
Analyzing 12 replicates (water spiking at 22 fg g−1 of 226Ra), the 226Ra recovery efficiency close to 100 % has been obtained thanks to the optimization of the radium separation method and a pre-concentration of radium in groundwater samples. The same could be said with respect to the separation of the interfering elements, important for the quantitative 226Ra analysis by SF-ICP-MS. The improvements in the separation and pre-concentration techniques also helped to improve the 226Ra detection limit down to 0.05 fg/25 mL of groundwater sample. The absolute detection limit for a water sample equivalent of 1.95 10−18 g g−1 has been obtained.
We presented an optimization of the SF-ICP-MS technique for the analysis of 226Ra in groundwater samples using a method of pre-concentration of radium in water samples. The separation protocol and a sequential application of ion exchange and extraction chromatography have been optimized, and related polyatomic interferences and matrix effects affecting the 226Ra signal were investigated. The 226Ra recovery efficiency close to 100 % has been obtained thanks to the optimization of the radium separation method and a pre-concentration of radium in groundwater samples. The same could be said with respect to the separation of the interfering elements, important for the quantitative 226Ra analysis by SF-ICP-MS.
The improvements in the separation and pre-concentration techniques helped to improve the 226Ra detection limit down to 0.05 fg/25 mL of groundwater sample. The absolute detection limit for a water sample is equivalent to 1.95 10−18 g g−1. A hydrological paper describing the measured 226Ra concentrations and discussing its behavior in the groundwater system of the Gran Sasso massif is under preparation.
The authors greatly acknowledge the support by the National Scientific Committee Technology of the National Institute of Nuclear Physics for the ERMES project and the Chemistry and Chemical Plants Service of the Gran Sasso National Laboratory.
- Butler OT, Cairns WRL, Cook JM, Davidson CM (2015) 2014 atomic spectrometry update—a review of advances in environmental analysis. J Anal At Spectrom 30:21–63View ArticleGoogle Scholar
- Chabaux F, Othman D, Birck J (1994) A new Ra-Ba chromatographic separation and its application to Ra mass-spectrometric measurement in volcanic rocks. Chem Geol 114:191–197View ArticleGoogle Scholar
- Cohen AS, O’Nions RK (1991) Precise determination of femtogram quantities of radium by thermal ionization mass spectrometry. Anal Chem 63:2705–2708View ArticleGoogle Scholar
- Cozzella ML, Leila A, Hernandez RS (2011) Determination of 226Ra in urine samples by Q-ICP-MS: a method for routine analyses. Radiat Meas 46:109–111View ArticleGoogle Scholar
- Epov V, Lariviere D, Evans R, Li C, Cornett R (2003) Direct determination of 226Ra in environmental matrices using collision cell inductively coupled plasma mass-spectrometry. J Radioanal Nucl Chem 256:53–60View ArticleGoogle Scholar
- Hamester M, Wiederin D, Wills J, Kerl W, Douthitt CB (1999) Strategies for isotope ratio measurements with a double focusing sector field ICP-MS. Fresen J Anal Chem 364:495–498View ArticleGoogle Scholar
- Horwitz EP, Chiarizia R, Dietz ML (1992) A novel strontium-selective extraction chromatographic resin. Solvent Extr Ion Exc 10:313–336View ArticleGoogle Scholar
- Hou X, Roos P (2008) Critical comparison of radiometric and mass spectrometric methods for the determination of radionuclides in environmental, biological and nuclear waste samples. Anal Chim Acta 608:105–112View ArticleGoogle Scholar
- Johnston A, Martin P (1997) Rapid analysis of 226Ra in waters by gamma-ray spectrometry. App Radiat Isot 48:631–638View ArticleGoogle Scholar
- Kim G, Burnett W, Dulaiova H, Swarzenski P, Moore W (2001) Measurement of 224Ra and 226Ra activities in natural waters using a radon-in-air monitor. Environ Sci Technol 35:4680–4683View ArticleGoogle Scholar
- Lariviere D, Epov VN, Evans RD, Cornett RJ (2003) Determination of radium-226 in environmental samples by inductively coupled plasma mass spectrometry after sequential selective extraction. J Anal At Spectrom 18:338–343View ArticleGoogle Scholar
- Lariviere D, Epov VN, Reiber K, Cornett R, Evans R (2005) Micro-extraction procedures for the determination of Ra-226 in well waters by SF-ICP-MS. Anal Chim Acta 528:175–182View ArticleGoogle Scholar
- Lee SH, Povinec PP, Wyse E, Hotchkis MAC (2008) Ultra-low-level determination of 236U in IAEA marine reference materials by ICPMS and AMS. Appl Radiat Isot 66:823–828View ArticleGoogle Scholar
- Lehto J, Hou X (2010) Chemistry and analysis of radionuclides: laboratory techniques and methodology. Wiley, New York, p 426View ArticleGoogle Scholar
- Lujanienė G, Beneš P, Štamberg K, Jokšas K, Kulakauskaitė I (2013) Pu and Am sorption to the Baltic Sea bottom sediments. J Radioanal Nucl Chem 295:1957–1967View ArticleGoogle Scholar
- Morvan K, Andres Y, Mokili B, Abbe JC (2001) Determination of radium-226 in aqueous solutions by alpha-spectrometry. Anal Chem 73:4218–4224View ArticleGoogle Scholar
- Park CJ, Oh PJ, Kim HY, Lee DS (1999) Determination of 226Ra in mineral waters by high-resolution inductively coupled plasma mass spectrometry after sample preparation by cation exchange. J Anal At Spectrom 14:223–227View ArticleGoogle Scholar
- Plastino W, Povinec PP, De Luca G, Doglioni C, Nisi S, Ioannucci L, Balata M, Laubenstein M, Bella F, Coccia E (2010) Uranium groundwater anomalies and L'Aquila earthquake, 6th April 2009 (Italy). J Environ Radioactiv 101:45–50View ArticleGoogle Scholar
- Plastino W, Panza GF, Doglioni C, Frezzotti ML, Peccerillo A, De Felice P, Bella F, Povinec PP, Nisi S, Ioannucci L, Aprili P, Balata M, Cozzella ML, Laubenstein M (2011) Uranium groundwater anomalies and active normal faulting. J Radioanal Nucl Chem 288:101–107View ArticleGoogle Scholar
- Plastino W, Laubenstein M, Nisi S, Peresan A, Povinec PP, Balata M, Bella F, Cardarelli A, Ciarletti M, Copia L, De Deo M, Gallese B, Ioannucci L (2013) Uranium, radium and tritium groundwater monitoring at INFN-Gran Sasso National Laboratory, Italy. J Radioanal Nucl Chem 295:585–592View ArticleGoogle Scholar
- Repinc U, Benedik L (2002) Development of a method for the determination of 226Ra by liquid scintillation counting. J Radioanal Nucl Chem 254:181–185View ArticleGoogle Scholar
- Reynolds BC, Wasserburg GJ, Baskaran M (2003) The transport of U- and Th-series nuclides in sandy confined aquifers. Geochim Cosmochim Acta 67:1955–1972View ArticleGoogle Scholar
- Smith C, Swarzenski P, Dimova N, Zhang J (2012) Natural radium and radon tracers to quantify water exchange and movement in reservoirs. Handbook of Environmental Isotope Geochemistry. Springer, Berlin, Heidelberg, pp 345–365Google Scholar
- Varga Z (2008) Ultratrace-level radium-226 determination in seawater samples by isotope dilution inductively coupled plasma mass spectrometry. Anal Bioanal Chem 390:511–519View ArticleGoogle Scholar
- Wyse EJ, Lee SH, La Rosa J, Povinec P, de Mora SJ (2001) ICP-sector field mass spectrometry analysis of plutonium isotopes: recognizing and resolving potential interferences. J Anal Atom Spectrom 16:1107–1111View ArticleGoogle Scholar
- Zheng J, Yamada M (2006) Determination of U isotope ratios in sediments using ICP-QMS after sample cleanup with anion-exchange and extraction chromatography. Talanta 68:932–939View ArticleGoogle Scholar
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.