Simultaneous determination of 210 Pb and 90 Sr and 210 Po isolation in sludge samples using a plastic scintillation resin

This study describes a new and fast method for separating 210 Po from 210 Pb and 90 Sr, before simultaneously measuring the individual activities of the latter two radionuclides using a plastic scintillation resin (PSresin) in sludge samples taken from a drinking water treatment plant. This method speeds up the analysis process significantly by simultaneously measuring 210 Pb and 90 Sr in a single step. The method is reproducible and has a relative standard deviation of less than 25% for 210 Pb, 210 Po and 90 Sr. The method was satisfactorily validated with an intercomparison sample and applied to sludge samples from a drinking water treatment plant. The minimum detectable activities for 0.9 g of sludge are 5.5 Bq/kg and 8 Bq/kg


Introduction
The drinking water treatment plant (DWTP) located in L'Ampolla (Tarragona) collects water from the Ebro River and distributes it across the province of Tarragona.Besides the influence of the terrain's lithology (Martínez et al., 2021b), the main radioactive influences in the lower basin of the Ebro River are the dicalcium phosphate (DCP) production plant, a naturally occurring radioactive material (NORM) industry, and a nuclear power plant (NPP) (Palomo et al., 2010).Generally, the gross alpha and gross beta levels in the water that feeds the DWTP are not significantly high (Palomo et al., 2010) and therefore, the water distributed to the population following the treatment processes comply with Spanish Royal Decree 314/2016 (RD 314/2016(RD 314/ , 2016)).However, some of the radionuclides that are present in the water stream at very low concentrations can be concentrated in the sludge, generated as a by-product of the different water treatment processes in the DWTP (Peñalver et al., 2020).This matrix can be useful for assessing the presence of some radionuclides that would otherwise be very difficult to measure directly in water samples due to their low levels.
In previous studies focused on determining the presence of different radionuclides in sludge produced by the DWTP, our laboratory found that the main gamma-emitting radionuclides were 4 K, 7 Be and 238 U and 232 Th and their decay chain daughters (Palomo et al., 2010).Also, among the descendant isotopes of 238 U, the beta emitter 210 Pb (Mola et al., 2014) and the alpha emitter 210 Po were also determined, with average activities of 37 ± 5 Bq/kg (Fonollosa et al., 2015).The presence of the previously mentioned radionuclides is mainly linked to the lithology of the river basin (Martínez et al., 2021b).Moreover, the presence of some artificial beta radionuclides such as 63 Ni and 55 Fe (with average activities of 45.5 Bq/kg and 112 Bq/kg respectively) (Fonollosa et al., 2016) as well as 90 Sr (with an average activity of 55 Bq/kg) (Mola et al., 2014) was determined and in this case, their origin could be attributed to the NPP located upstream of the aforementioned DWTP.
In general, determining radionuclides requires radiochemical sample pretreatments, which tends to be a complex, tedious and lengthy process (Grate and Egorov, 2003).However, methods that are able to provide a fast response are needed, particularly for emergency situations.Among all the radionuclides we determined in previous studies in sludge samples from the DWTP in L'Ampolla, the present study focuses on determining 210 Pb, 210 Po and 90 Sr due to their radiotoxicity and because they are common radionuclides in natural waters, especially those influenced by NPPs in the case of 90 Sr presence (International Atomic Energy Agency, 1963).In this regard, the literature reports several methods using an extraction chromatographic Sr resin to isolate one or even two of these three radionuclides prior to their measurement, in different kind of matrices.Vajda et al. (1997) proposed the sequential analysis of 210 Pb and 210 Po using this resin and since then, it has usually been the method of choice for isolating these radionuclides in environmental matrices (Kong et al., 2021a;S ¸ahin and Özbay, 2021;Vreček and Benedik, 2002).Given that the time it takes to complete the analysis is one of the most critical considerations when emergency situations arise, our laboratory recently developed a rapid method to sequentially separate 210 Pb, 210 Po and 90 Sr in water samples (Martínez et al., 2021a), as these are targeted radionuclides that must be analysed in drinking water according to Spanish Royal Decree 314/2016 (RD 314/2016(RD 314/ , 2016)).In this regard, the literature also mentions some other examples also focused on accelerating and optimising the analytical methodology (Fons-Castells et al., 2017a;Mola et al., 2014).Some approaches based on automating the separation procedure, such as those reported by Mola et al. (2014), focused on the determination of 90 Sr and 210 Pb in DWTP sludge through lab-on-valve multisyringe flow injection analysis (LOV-MSFIA).This has been proven to be a good alternative to traditional radiochemical methods as it reduces reagent consumption and minimises the amount of resin used, and the effluents generated.Other approaches use mathematics to simultaneously and quickly determine the individual activities of different radionuclides present in a given sample (Bagán et al., 2011a;Fons-Castells et al., 2017a,b), as well as some that simplify the conventional separation and sample preparation.
In recent years, a new family of extraction resins with scintillating properties has been developed by the University of Barcelona, named plastic scintillation resin (PSresin) (Barrera et al., 2016;Giménez et al., 2021).This new type of resin is capable of selecting each radionuclide being studied, as it is composed by selective extractants.In this regard, a PSresin for strontium determination was designed as an analogy of the Sr resin (TrisKem) and the plastic scintillator microspheres were coated in DtBuCH18C6 in 1-octanol.The use of this PSresin offers different advantages to Sr resin, such as reducing the time, labour and reagents needed for analysis, unifying chemical separation and sample measurement so they can be completed in one step, and directly measuring radionuclides using liquid scintillation counting (LSC).In the few last years, the PSresin for strontium has been used to determine 210 Pb, 89 Sr and 90 Sr. Lluch et al. (2016) take advantage of its high selectivity for 210 Pb to develop a method to determine this radionuclide by combining separation and measurement preparation into a single step, thus reducing the reagents required for analysis and the amount of organic waste generated.In the same vein, Sáez-Muñoz et al. (2019, 2018) report two rapid methods for determining the individual activities of 89 Sr and 90 Sr in aerosol filters and vegetation using deconvolution methods and total radiostrontium ( 89 Sr+ 90 Sr) in milk.Both are robust and reproducible methods that could be used in emergency situations given the significant reduction in time needed for analysis.
In view of this, this study's main objective is to develop a novel method capable of determining the individual activities of 210 Po, 210 Pb and 90 Sr in sludge samples by using a PSresin.To achieve this, we will isolate the 210 Po, with its subsequent alpha measurement, then simultaneously measure the 210 Pb and 90 Sr attached to the resin using direct scintillation counting.This method would significantly reduce the time needed to analyse the individual activities of 90 Sr, 210 Pb and 210 Po while also reducing the number of reagents used in comparison with the methods previously described.

Reagents and materials
All reagents were of analytical or reagent grade.Hydrochloric acid (37%) was supplied by Merk (Darmstadt, Germany), nitric acid (65%) was supplied by AvantorTM (Gliwice, Poland), lithium nitrate (>95%) was supplied by Chem-Lab NV (Zedelgem, Belgium), strontium nitrate (>99%) was supplied by Fluka Chemie (Buchs, Switzerland), lead nitrate (>99%) was supplied by Probus (Spain) and scintillation cocktail Optiphase HiSafe 3 was supplied by PerkinElmer™ (Waltham, USA).Ultrapure water was obtained from a Purification System Milli-Q Reference.The silver disks onto which 210 Po was autodeposited were supplied by Goodfellow (Huntingdon, United Kingdom), the twelve-hole vacuum box were supplied by TrisKem International (France).The prepacked PSresin cartridges (1.5 g of PSresin in 2 mL solid-phase extraction cartridges) were provided and produced by the Department of Chemical Engineering and Analytical Chemistry of the University of Barcelona.They are composed of 60 μm plastic scintillation microspheres coated by DtBuCH18C6 in 1-octanol in a proportion PSm: extractant of 1:1/8 and prepared as the authors previously described (Bagán et al., 2011b).A twelve-position vacuum box from TrisKem International (Rennes, France) was used.
The following stock solutions were used to prepare the calibration samples.A 210 Pb stock solution of 1545.93 (31.86)Bq/kg in a 3 M HNO 3 medium was prepared from a standard of 0.58 (0.01) Bq/mg and a 90 Sr stock solution of 1764.98 (33.73)Bq/kg in a 3 M HNO 3 medium was prepared from a standard of 0.123 (0.001) Bq/mg, both provided by Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas (CIEMAT) (Spain).A 209 Po stock solution of 4989 (748) Bq/kg was prepared from standard of 100 Bq/mL provided by Eckert & Ziegler (United States).The values in brackets refer to the standard uncertainty.

Certified reference materials and samples
To validate the method, a certified reference material obtained from an intercomparison exercise organised by the Spanish Nuclear Council (CSN) was used (CSN/CIEMAT-05).The certified reference material was made from wood ash and contained 90 Sr and 210 Pb with the following respective reference activities, 91.2 ± 1.4 Bq/kg and 102.0 ± 7.4 Bq/kg.This sample also contained 210 Po, the activity of which was considered equal to 210 Pb as they were in secular equilibrium.
The presence of 90 Sr, 210 Pb and 210 Po was also evaluated in a sludge sample from the DWTP located in L'Ampolla.This was a mixture of single samples collected once a week over the course of a month in a polyethylene bottle.
It is important to highlight that the certified reference material and the real sample analysed (sludge) are not similar in terms of composition.Even that, and due to the lack of a certified sludge containing the three radionuclides under study, the wood ash reference material has been chosen to validate the present method taking into account the possible limitations using this approach.

Equipment and software
90 Sr and 210 Pb were measured using a 1220 Quantulus™ LS counter supplied by PerkinElmer™ (Waltham, USA), which was operated using the software WinQ version 1.3.The data obtained was viewed and treated using Matlab R2020b.Alpha measurements were performed using an Octête plus™ alpha detector supplied by Ortec (Oak Ridge, United States) and the data obtained was viewed using Maestro.Analysis of stable elements were performed with a 7500ce inductively coupled plasma mass spectrometer (ICP-MS) and octopole reaction system supplied by Agilent (Santa Clara, United States), as well as an optima 8300 ICP-OES spectrometer supplied by PerkinElmer™.A Digital Orbital Shaker supplied by Heathrow Scientific (Vernon Hills, United States) was used to shake the vials for the batch experiments.The samples were digested using a Speedwave® Four microwave digestion system with built-in, non-contact temperature and pressure measurement supplied by Berghof (Germany).

Batch studies
First, it was necessary to determine the optimal composition and molarity of the acid solution in order to condition and load the sample onto the PSresin for the radionuclides being studied.The procedure used was as follows.First, 1 mL of solution containing 0.1 mL of Sr 2+ and Pb 2+ solutions (10,000 ppm, respectively) and 0.5 mL of 209 Po (5000 Bq/L) at the appropriate HNO 3 and HCl mixture molarity (final concentrations of 3 M and 1 M, respectively) was mixed with 1 mg of PSresin.The vials were shaken with an orbital shaker for 24 h and the resulting solution was filtered using a syringe and 0.45 μm cellulose filter.The Sr 2+ and Pb 2+ present in the acidic phase were determined by ICP (OES and MS), while 209 Po activity was determined by LSC.

Final analytical procedure
A sample pretreatment method based on a microwave digestion was chosen according to previous studies performed in our laboratory (Fonollosa et al., 2015;Mola et al., 2014).Calcination was avoided to prevent the volatilisation of 210 Po (Thakur and Ward, 2020).For this purpose, 0.9 g of sludge were transferred into a vessel with 15 mL of a mixture of HNO 3 :HCl (3:1), 1 mL of Pb 2+ and Sr 2+ solutions (1000 ppm, respectively) and 2 mL of 209 Po (19 ± 2 Bq/kg).Microwave digestion was then performed for 30 min by increasing the temperature from 0 to 140 • C over 5 min, and then to 175 • C over 25 min.The vessel was kept in the fridge overnight to prevent losing polonium.The next day, the sample was filtered using a 45 μm pore size syringe filter to remove any potential insoluble particles.The soluble part was then evaporated until it was dry at a controlled temperature of 80 • C and then finally redissolved to a final volume of 20 mL with a mixture of HNO 3 and HCl with a final concentration of 3 M and 1 M, respectively.
The sequence for isolating and preconcentrating 90 Sr, 210 Pb and 210 Po was performed following the steps described in Table 1.
First, the resin was pre-conditioned with 2 mL of the acidic solution (3 M HNO 3 and 1 M HCl).Then, 20 mL of digested sample was loaded into the cartridge.
Next, 210 Po was eluted with two fractions of 10 mL of 8 M HNO 3 and the cartridge was rinsed with 4 mL of 6 M LiNO 3 to reduce/avoid chemiluminescence when it came to measurement, due to the presence of nitric acid in the cartridge as it showed good results in other studies on PSresin (Sáez-Muñoz et al., 2018, 2019).The 210 Po present in each eluate was autodeposited on a silver disk.Soon after, autodeposition was performed for 3 h at 85 • C in an 80 mL 2 M HCl solution, containing 0.1 g of ascorbic acid (to reduce Fe 3+ and avoid its interference during the deposition process), whilst being continuously stirred, and then counted using an alpha spectrometer (Fonollosa et al., 2015;Martínez et al., 2021a).Then, the PSresin was directly measured for 3 h with a 1220 Quantulus™ LS counter as soon as the separation was performed to minimise the growth of daughter radionuclides from 210 Pb and 90 Sr. Elution fractions and waste were analysed by ICP to evaluate possible losses of 90 Sr and 210 Pb.The retention efficiency of 210 Pb and 90 Sr was evaluated by the determination of stable Pb and Sr by ICP-OES and ICP-MS, respectively, and calculated by determining the initial concentration and final concentration after PSresin isolation.The 210 Po overall method recovery was evaluated by determining the initial and final activity of 209 Po by alpha spectrometry.
The developed method shares some similarities with a previously published method in our laboratory (Martínez et al., 2021a) focused on the fast separation and determination of the same radionuclides in water.However, and even the similarities, the presented reported strategy here constitutes a further step in which new developed resins are employed.The use of these resins is related to greener strategies as these minimise the waste generation and unifies separation and measure steps.

Calibration
Three replicates of deionised water samples with known amounts of 90 Sr and 210 Pb (150 Bq/kg) were processed with a PSresin as described in section 2.4.1.They were measured in the liquid scintillator counter in order to calibrate the detection efficiencies for each appropriate window (for 90 Sr, a window between channels 375 and 650 and for 210 Pb, a window between channels 125 and 325).Also, two replicates deionised water samples containing a mixture of 90 Sr and 210 Pb with different ratios (1:1, 1:2 and 2:1) were processed to optimise the windows setting.In all cases, 0.5 mL of stable Sr 2+ and Pb 2+ solutions (10,000 ppm respectively), were added to determine the retention efficiency.

Data treatment
The retention efficiency (%R) of each radionuclide was calculated for each step of the separation using equation ( 1 where: the 90 Sr CPMs (counts per minute) of the sample and those corresponding to the blank were measured in a window between channels 375 and 650, the sample amount (kg), the detection efficiency (Eff 375-600 ) is 0.658 (0.1) and R is 90 Sr's retention efficiency of the sample (both given as a unit fraction), and f is the decay correction factor.
To determine the activity of 210 Pb, the corresponding 90 Sr CPMs interfering in 210 Pb's window (between channels 125 and 325) must be calculated first using equation ( 3 where: the 210 Pb CPMs, the 90 Sr CPMs and those corresponding to the blank were measured in a window between channels 125 and 325, the sample amount (kg), the detection efficiency (Eff 125-325 , 210 Pb's windows) is 0.601 (0.1) and R is 210 Pb's retention efficiency of the sample (both given as a unit fraction), and f is the decay correction factor.

Table 1
Sequential conditions for 210 Po separation and for 90 Sr and 210 Pb measurement using the PSresin. Step

Results and discussion
The following section discusses the optimisation processes that were performed to retain 90 Sr, 210 Pb and 210 Po in the resin when loading the sample, and to selectively elute 210 Po.It also describes calibrating the detection efficiency of 90 Sr and 210 Pb.Finally, it presents the validation of the method using reference material, as well as its applicability to real samples.

Batch study of 90 Sr, 210 Pb and 210 Po in PSresin
Initially, the acidic conditions used to condition and load the sample (10 mL of 3 M HNO 3 ) in the PSresin were the same as those used in one of our previous studies based on a sequential separation of 90 Sr, 210 Pb and 210 Po in drinking water samples using a Sr resin (TrisKem) (Martínez et al., 2021a).Despite the analogy between both resins, those conditions do not work with PSresin.The results were unexpected.Low retention of 210 Po was found when loading the sample (22%) while 210 Pb and 90 Sr were retained well (99% for both radionuclides).To improve 210 Po's retention, a batch study was performed.The acidic solutions to condition and load samples (a mixture of 3 M HNO 3 and different concentrations of HCl in the range from 1 to 5 M) were selected after looking for advice in various studies.It should be noted that the studies we consulted are only based on Sr resin (TrisKem) because of their similarities and the lack of information and methods using PSresin.In this regard, some reported strategies demonstrate that 210 Po and 210 Pb are well retained in Sr resin at moderately low concentrations of HCl, while these conditions for retaining 90 Sr in the resin are not favourable (Vajda et al., 1997).However, when using HNO 3 :HCl mixtures at ratios of above 50:50%, the retention of 90 Sr radionuclide improved considerably (McLain et al., 2018).Although, as far as we know, in the reported literature there is no data related to 210 Pb and 210 Po's retention in HNO 3 :HCl mixtures in Sr resin.In general, the retention of the aforementioned radionuclides in Sr resin is characterised in pure mineral acids.
The results obtained from this study are shown in Fig. 1, where the percentages of retention for each radionuclide at a mixture of 3 M HNO 3 and different concentrations of HCl from 1 to 5 M can be observed.It shows that 210 Po presented a slight decrease in retention when a higher hydrochloric acid concentration was used.For this radionuclide, a retention value of 96% was obtained for an HCl concentration of 1 M and approximately 80% at an acid concentration of 5 M.This is in line with the bibliographic review by Thakur and Ward (2020) who reported that the retention of polonium is high across a range of HCl molarities, with 2 M HCl being the most typical Sr resin load solution.
As regards lead and as shown in Fig. 1, this element presented a significant decrease in retention when a higher hydrochloric concentration was used.For this radionuclide, a retention value of 99% was obtained for an HCl concentration of 1 M and approximately 5% at an acidic concentration of 5 M.These results are in line with some studies stating that this radionuclide is retained well in HCl (0.1-2 M) and Sr resin (Eichrom Technologies, 2021;Kong et al., 2021b). 90Sr was retained well when a lower hydrochloric concentration was used.A retention value of approximately 88% was obtained for an HCl concentration of 1 M.However, the affinity of 90 Sr with PSresin as well as Sr resin reaches maximum values at between 3 and 8 M HNO 3 (Dianu and Dobrin, 2020), as is the case in the present study.
Finally, considering all the results and retention percentages for the three radionuclides involved in this study, the mixture chosen to load the sample and condition PSresin in the separation procedure was the 3 M HNO 3 and 1 M HCl final molarities mixture.This was a compromise between all the values obtained and thus ensured the optimal retention of each radionuclide in the PSresin.Therefore, mixed nitric and hydrochloric acid matrix are useful to perform separations in PSresin.

Optimisation of separation parameters to avoid interferences
Before measuring 90 Sr and 210 Pb in PSresin using liquid scintillation counting, potential interferences present in the sample must be removed as they could influence the separation and measurement of the radionuclides (Barrera et al., 2016;Giménez et al., 2021;Kocadag et al., 2013;Sáez-Muñoz et al., 2018).For example, if 210 Po cannot be completely stripped from PSresin, it could interfere with the quantification of 90 Sr and 210 Pb as it appears approximately between channels 300 and 600 (Kim et al., 2008;Lluch et al., 2016).To achieve this, an optimisation of the necessary volume of 8 M HNO 3 was carried out, which is the most common acidic media used in the literature to strip 210 Po from Sr resin (Hurtado-Bermúdez et al., 2017).Using this elution media means that 90 Sr and 210 Pb do not coelute with 210 Po.The first is retained well in nitric acid concentration by the Sr resin (Triskem International, 2015) and 210 Pb elutes with 0.06 M ammonium oxalate (Martínez et al., 2021a;Mola et al., 2014).
Different volumes of 8 M HNO 3 , ranging from 8 to 20 mL, were used to try to perform the quantitative elution of polonium.As shown in Fig. 2, 60% of 210 Po was eluted by using 8 mL of 8 M HNO 3 , which increased to approximately 100% when 20 mL was used.The latter, therefore, has proven to be the optimum volume to quantitatively elute almost all polonium retained in PSresin, thus avoiding radiochemical interferences when measuring 90 Sr and 210 Pb using LSC.Moreover, a PSresin rinse with 4 mL of LiNO 3 was also applied to reduce and avoid chemiluminescence.

Calibration of 90 Sr and 210 Pb detection efficiency
The next step was to measure both the 90 Sr and 210 Pb retained in the resin by using a LSC spectrometer.When performing this quantification and as shown in Fig. 3, it is important to consider that the spectra of both radionuclides overlap in some regions, which can cause activities to be overestimated.Different methods for LSC spectra deconvolution that avoid overlapping are found in the literature (Sáez-Muñoz et al., 2019), and among them, partial least squares (PLS) have been used to determine alpha and beta emitters in mixtures (Fons-Castells et al., 2017a).In this case, Fons-Castells et al., 2017a apply a multivariate calibration method using PLS regression to avoid overlap in the spectra between 228 Ra and 210 Pb.In the present study and in order to avoid or minimise 90 Sr and 210 Pb spectra overlapping, we used a calibration model by windows.It involved a two-window approach, one for 90 Sr and one for 210 Pb.Data from the standards was used to select the optimum count windows, compromising between the detection efficiency for each radionuclide, as follows.
Individual standards ( 90 Sr and 210 Pb) were processed following the procedure explained in section 2.4.1.The corresponding CPMs and retention efficiencies could be obtained from the PSresin measurement.The CPMs of the individual standard of 90 Sr were measured in ranges of more than 100 channels between channels 300 and 650.The detection efficiency was measured for each window using equation (4).

Detection efficiency
The 90 Sr CPMs were calculated for each mixture of standards ( 90 Sr + 210 Pb) using the previously calculated detection efficiencies and the calculated CPMs were compared with the theoretical values using equation ( 5) to evaluate the error of the method (%).The optimal window was considered the one for which detection efficiency led to the lower error in the CPM calculations.

% Error
The same procedure was performed for the individual standards between channels 0 and 350 to determine both 90 Sr and 210 Pb's detection efficiency in 210 Pb's window.The respective CPMs of 210 Pb and 90 Sr in the mixtures of standards were calculated, and their errors were calculated using equation ( 5).As shown in Table 2, the optimum windows were from channels 375 to 600 to quantify 90 Sr, and from channels 125 to 325 to quantify 210 Pb.The related optimal detection efficiencies shown are in line with those obtained in studies in the literature determining 90 Sr and 210 Pb using PSresin (Bagán et al., 2011b;Lluch et al., 2016).
The calibration procedure was validated by quantifying three water samples containing different ratios of 90 Sr and 210 Pb (1:1, 1:2 and 2:1).The deviations obtained when quantifying 90 Sr and 210 Pb are shown in Table 3.The relative bias of 90 Sr and 210 Pb is less than 15% in all three cases, so the approach can be used to quantify 90 Sr and 210 Pb simultaneously in one sample.It is important to note that when the concentration of 90 Sr is two times higher than the concentration of 210 Pb, the relative bias percentage of the latter is high (14.1%).This is due to the fact that there is more 90 Sr interfering in channels 150 to 325 and 210 Pb activity could be overestimated.Therefore, the present mathematical method is a useful tool for determining individual 90 Sr and 210 Pb in emergency situations, taking into account that 210 Pb activities could be overestimated when there are higher levels of 90 Sr in the sample due to how it can interfere in the 210 Pb spectra.

Method validation and application to real samples
The developed method was validated by analysing certified reference material (CSN/CIEMAT-05) made from wood ash.The sample contained 90 Sr and 210 Pb, with 91.2 and 102 Bq/kg of dry mass, respectively, at the time of separation. 210Pb and 210 Po were in secular equilibrium.As shown in Table 4, the mean values of the three replicates obtained generally agreed with the certified values, as Z-score (<1) indicates.Regarding the accuracy of the measurements, the average relative standard deviation of each radionuclide is low (<10%) with the exception of 90 Sr, which shows a deviation that is higher than expected.This could be related to the variation of 90 Sr PSresin retention due to changes in the acid molarity mixtures used for the resin conditioning and sample loading.In this regard, it is worth mentioning the published work of McLain et al. (2018) who reported that mixtures of HNO 3 and HCl can evolve to their decomposition into volatile gaseous products  and their initial concentration can decrease and this can influence on the retention of the radionuclides.
After the validation procedure, the method developed in this study was applied to three replicates of a sludge sample from the DWTP located in L'Ampolla.The results obtained are displayed in Table 5.
It shows that 210 Pb presented an averaged activity (n = 3) of 23.9 ± 4.9 Bq/kg.This sludge sample was also analysed using gamma spectrometry to compare the 210 Pb results with those obtained in the present study, obtaining a result of 21.2 ± 4.9 Bq/kg, which, by value comparison, indicates a satisfactory performance of the method in terms of lead quantification.This radionuclide was previously quantified using a LOV-MSFIA automated method in our laboratory, which obtained activities in a range of 38-63 Bq/kg (Mola et al., 2014).It is important to note that the results obtained in the present study are slightly lower than those obtained by Mola et al. (2014) in sludge samples from the same DWTP collected in 2014.The variability of these results could be due to seasonal changes in the Ebro River system, behaviour that has been observed in other natural radionuclides (Martínez et al., 2021b).
As regards 90 Sr, the average activity obtained is 17.8 ± 5.3 Bq/kg.This activity is lower than that obtained in a previous study performed in our laboratory (34-78 Bq/kg) (Mola et al., 2014).As DWTP sludge concentrates radionuclides present in the influent (from the Ebro River) at the DWTP, it can be concluded that the presence of 90 Sr in the sludge could be related to the normal operational radioactive liquid discharges found in the Ebro River from Ascó NPP, located upstream, as reported by the Spanish Nuclear Council (CSN) (P.Martínez et al., 2021).

210
Po presented an activity of 25.5 ± 4.9 Bq/kg, which is comparable with 210 Pb activity because both were in secular equilibrium.Fonollosa et al. (2015) determined the presence of polonium in this DWTP's sludge years ago, by means of sample digestion and direct autodeposition onto silver disks, in the range of 31-42 Bq/kg.It is a particle-reactive radionuclide that tends to be associated more with the particulate phase and consequently is found concentrated in sludge (Fonollosa et al., 2015).According to UNSCEAR (2008), the obtained activity values are quite comparable to the 210 Po activities expected in normal soils and it seems that the DCP production plant located upstream does not have any influence.
Table 5 also shows the minimum detectable activity (MDA) calculated in the present study using the Currie approach (Currie, 1968) in a sample of 0.9 g.The current method shows improvements in the MDAs of 90 Sr and 210 Pb compared to other methods reported in the literature (Mola et al., 2014;Qiao et al., 2017).Mola et al. (2014) developed an automated method based on LOV-MSFIA using Sr resin and LSC, where the MDAs for 90 Sr and 210 Pb were 15 and 13 Bq/kg respectively for a sludge sample of 0.5 g that was measured for 500 min.On the other hand, Qiao et al. (2017) report a MDA for 90 Sr of 10.4 Bq/kg by analysing 1 g of soil and measuring it for 2400 min through 90 Y.In the case of 210 Po, the MDA obtained remains in the same order of magnitude as the conventional methods used in our laboratory (0.15 Bq/kg for 0.3 g of sample and measured for 5833 min) (Fonollosa et al., 2015).They are also in line with the MDA value obtained by El Afifi and Borai (2006) (0.28Bq/kg), who determined 210 Po by using Sr resin, a 0.3-0.7 g soil sample after measuring it for 1440 min.

Conclusions
This study has successfully developed a fast and novel method to determine 210 Po, 210 Pb and 90 Sr using a PSresin by selectively retaining the three radionuclides, eluting 210 Po and subsequently measuring it by alpha spectrometry, while measuring the 210 Pb and 90 Sr retained in the PSresin by LSC.The optimal loading medium was found to be a mixture of 3 M HNO 3 and 1 M HCl and the optimal elution method for 210 Po was 20 mL of 8 M HNO 3 .As the PSresin combines the steps involving separation and measurement, less time is needed for analysis.Finally, the mathematical method proposed and tested to determine 90 Sr and 210 Pb simultaneously provided good results.
The method was successfully validated by analysing reference material and obtaining activities statistically similar to the reference values.It was then applied to a real sludge sample from the DWTP located in L'Ampolla (Tarragona).
In conclusion, it has been shown to be a successful method for determining the aforementioned radionuclides in rapid-response situations.

Table 4
Validation results obtained in three replicates (n = 3) of the CSN/CIEMAT-05 intercomparison sample.Uncertainties are expressed at k = 2.

Table 5
Results obtained from the analysis of a sludge sample from the DWTP.
mass of the tracer (mg), both in the waste and in the sample, were determined by ICP-OES in case of stable Pb and by ICP-MS in case of stable Sr.Equation (2) was used to determine the activity of 90 Sr: Activity 90 Sr = 90 Sr CPM 375− 650 − blank CPM 375− 650 60 x sample amount x R x Eff 375− 650 x f (2) ): CPM90 Sr in 125− 325 = CPM90 Sr × Eff 125− 325 × R (3) where: the detection efficiency (Eff 125-325 ) is 0.195 and R is 90 Sr's retention efficiency of the sample (both given as a unit fraction).The activity of 210 Pb was calculated by subtracting the interfering 90 Sr CPMs to the total CPM measured in 210 Pb's window using equation (4): Activity 210 Pb = 210 Pb CPM 125− 325 − 90Sr CPM 125− 325 − blank CPM 125− 325 60 x sample amount x R x Eff 125− 325 x f (4a)

Table 2
Optimal spectrum windows for each radionuclide and their related detection efficiency.

Table 3
Relative bias in the determination of 90 Sr and 210 Pb in water samples with different activity ratios.
MDA calculated for 5000 min of measurement time.bMDA calculated for 180 min of measurement time.deGestió d'Ajuts Universitaris i de Recerca (AGAUR) for financial support under 2017-SGR-907.I. Giménez also thanks the University of Barcelona for the PREDOCS-UB grant. a