Introduction

Radionuclides 210Pb and its granddaughter radionuclide 210Po are classified as highly toxic radioisotopes, representing a significant source of internal natural radiation dose for humans. Several radioanalytical techniques are available for the measurement of 210Pb, each grounded in distinct physical and chemical principles [1]. These methods vary in terms of achievable detection limits, selectivity, analytical complexity, reproducibility, and resilience to diverse chemical compositions and varying levels of other natural radionuclides. Three commonly employed radiometric methods for 210Pb measurement in samples include gamma-ray spectrometry of 210Pb in various media such as water, rocks, soil, and sediment; beta counting and spectrometry, involving the observation of the growth of its daughter nuclide 210Bi; and alpha-particle spectrometry of its granddaughter 210Po, assuming radioactive equilibrium between the two radionuclides [1]. In the publication by Sakoda [2], the concentrations of 210Pb in tobacco samples were determined using gamma-ray spectrometry. Each sample, consisting of a few grams of tobacco, was enclosed in a 100 mL plastic container. Lead could be precipitated from the lead-containing eluate in the form of lead sulphate [3]. Afterwards, the precipitate was transferred to a counting plate and weighed to determine the amount of material recovered. The activity concentrations of 210Pb were determined by gamma-ray spectrometry. Gamma rays emitted from the samples were measured utilizing a high-purity germanium detector (GMX-15200, SEIKO EG&G, Japan). To determine the concentration of 210Pb, the gamma-ray spectrum was analysed for the presence of the 46.5 keV photopeak. The peak area was calculated using a non-linear least-squares fitting program that adopted Gaussian peak shapes. For both, sample and background measurements, the counting time was 86,400 s, and the detection efficiency for 210Pb was 4.25% [3, 4]. Peres and Hiromoto [5] employed the precipitation of Ba (Ra, Pb) SO4, and the use of PbS and PbCrO4 for the determination of 210Pb. The activity of 210Pb was determined by beta counting of 210Bi, which was grown in the final precipitate, using a low-background gas flow proportional counter. The efficiency of the counting process was estimated to be around 35%. The chemical yield of the process ranged from 61 to 98% and the minimum detectable concentration (95% confidence level) using this method was 4.5 m Bq/g with a counting time of 400 min. Savidou [6] measured the 210Pb activity in the samples using solutions prepared in accordance with the procedure outlined by Fisenne [7]. The activity concentration of 210Pb was determined by measuring the activity of 210Po. Fractions of polonium were stored for a duration of 5–7 months to enable the production of 210Po resulting from the decay of 210Pb over a known time interval. The chemical yield of the 210Po process ranged from 23 to 62% and the minimum detection activity of this method (95% confidence level) was 0.3 m Bq/g with a counting time of 1000 min. For the analysis of 210Pb, Din [8] utilized the residual solution from the initial deposition of 210Po, which was stored for a period exceeding 6 months to allow the growth of 210Po from 210Pb. Subsequently, 210Po was redeposited using 208Po as a yield tracer. The activity concentration of 210Pb was calculated using an equation from the publication by Vesterbacka [9]. A tracer recovery yield ranging from 60 to 85% was achieved and a minimum detectable activity of 1 m Bq was attained with a counting time of 1000 min.

Kovács [10] determined 210Pb indirectly by measuring 210Po in the samples after a delay of at least eight months. The minimum detectable activity (95% confidence level) achieved with this method was 0.67 mBq/g, and the counting time used was 80,000 s. A new sequential cloud point extraction procedure was developed by Blanchet [11] and validated for the determination of 210Pb and 210Po by ICP-MS/MS and alpha spectrometry, respectively. Two distinct cloud point extraction systems using 4′, 4″ (500)-di-tert-butyldicyclohexano-18-crown-6 and N,N,N′,N′-tetraoctyldiglycolamide as chelating agents were performed sequentially on the same samples. This analytical procedure has demonstrated outstanding rapidity, efficiency, versatility, and ease of execution. Sorbents for extraction chromatography, Eichrom’s Sr and Pb Resins are widely used for separating a variety of metal ions (including lead) in nitric acid media, valued for their efficiency. However, their capacity may diminish due to leaching of crown ether from the support during separation, which leads to reduced recovery and limited reuse. Our previous research has demonstrated the effectiveness of the molecular recognition sorbent AnaLig Sr01 in selectively concentrating and separating lead from bone samples. This material features a silica gel support with a covalently bound ligand, providing high specificity and the ability to separate ions in complex mixtures [12]. The aim of our current work is to apply this sorbent to the matrix of tobacco and determine the activity of 210Pb in tobacco and cigarettes commonly available in shops in Slovakia. The results will be presented in a separate publication. The primary objective of this study was to statistically compare the molecular recognition sorbent AnaLig Sr01 with the extraction chromatography sorbent Sr Resin for the determination of 210Pb in tobacco. To appropriately apply the separation and determination methodologies for 210Pb, it was essential to first statistically compare the suitability of both techniques. This paper focuses on the statistical comparison of the obtained results using linear regression, regression diagnostics, and a paired t test, to evaluate the AnaLig Sr01 separation method relative to the Sr Resin method.

Materials and methods

Materials

The methods for analysing 210Pb involved the use of a molecular recognition technology resin, AnaLig Sr01 (100–150 μm), available from IBC Technology (USA). The Sr Resin (50–100 μm) was supplied by TRISKEM International (France). A certified 210Pb standard (1000 mBq/g, relative uncertainty of 1.1%) was supplied by the Czech Metrology Institute (Prague, Czechia), with the reference date for all radioisotopes set to September 18, 2015. All other chemicals were of analytical grade.

Two types of chromatographic columns were used: Sr Resin sorbent (diameter: 0.5 cm; length: 15.5 cm) and for AnaLig Sr01 sorbent (diameter: 1.5 cm; length: 15.5 cm). The liquid scintillation cocktail Ultima GOLD AB and the liquid scintillation spectrometer TRI CARB 3100TR were employed. Atomic absorption spectrometers AAS 1100 and AAnalyst 400 (Perkin Elmer, USA) were employed. Other equipment included a hot plate (VWR, USA), a drying oven, scales (Kern, Germany) and analytical balances (Sartorius, Germany).

Methods

Twenty-five grams of tobacco samples were digested in 200 mL of 14.35 mol/L HNO3 with a constant stirring using a magnetic stirrer at a temperature of 140 °C for 9 h. The beaker was covered with a watch glass during digestion. The solution sample was filtered after cooling and then divided into twenty parts: ten parts for Sr Resin method and ten parts for the AnaLig Sr01 method. The solution was spiked with the radionuclide of 210Pb. A 250 µL standard solution of 210Pb with an activity of 1000 mBq/g and 0.2 mL of Pb2+ carrier with a concentration of 10 mg/mL were added to each sample. The sample concentrations were subsequently diluted with deionized water to achieve a final concentration of 8 mol/L HNO3.

Separation of 210Pb on Sr resin

On prepared chromatographic columns (dimensions 15.5 cm × 0.5 cm), 0.5 g of Sr resin sorbent was applied, the bed volume was 2.0 mL. The column with Sr resin sorbent was conditioned with 20 mL of 8 mol/L HNO3 at a flow rate of 1 mL/min. The sample was then loaded onto the column. Subsequently, the sorbent was washed with 15 mL of 8 mol/L HNO3. After elution with 8 mol/L HNO3, the sorbent was washed with 15 mL of deionized H2O. Lead was eluted with 15 mL of 9 mol/L HCL and the fraction was collected in a beaker. The solution was evaporated to a near dry-state at 180 °C. The residue was dissolved in 2 mL of 0.05 mol/L HNO3, quantitatively transferred to a vial to a final volume of 5 mL. Aliquots of 100 μL from the solution for Atomic Absorption Spectrometry were taken to determine the chemical yield of lead. Aliquots of 100 μL from the solution for Atomic Absorption Spectrometry were taken to determine the chemical yield of lead. These 100 μL aliquots were then diluted to a final volume of 10 mL.

Separation of 210Pb on AnaLig Sr01

Before applying AnaLig Sorbent to the column, it is advisable to let it swell in distilled water in a beaker for at least 30 min, during which the volume of the sorbent almost doubles. The prepared chromatographic columns were loaded with 0.5 g of AnaLig Sr01 sorbent, the bed volume was 2.6 mL. The column with the AnaLig Sr01 sorbent was conditioned with 20 mL of 8 mol/L HNO3 at a flow rate of 1 mL/min. The sample was applied to the column. AnaLig Sr01 was washed with 15 mL of 8 mol/L HNO3 and subsequently with 15 mL of deionized H2O. Lead was eluted with 15 mL of 9 mol/L HCL and the fraction was collected in a beaker. The solution was evaporated a near dry-state at a temperature of 180 °C. The evaporated solution was dissolved in 2 mL of 0.05 mol/L HNO3, quantitatively transferred into a vial to final volume 5 mL. Aliquots of 100 μL from the solution for Atomic Absorption Spectrometry were taken to determine the chemical yield of lead. These 100 μL aliquots were then diluted to a final volume of 10 mL.

Measurement of 210Pb on a liquid scintillation counter TRI CARB 3100TR

The chemical yield of lead was determined using an atomic absorption spectrometer AAnalyst 400/AAS 1100. Subsequently, 15 mL of scintillation solution ULTIMA GOLD AB was added, and 210Pb was measured on a liquid scintillation spectrometer TRI CARB 3100TR. An optimized energy window ranging from 50 to 700 keV was selected to minimize background counts and achieve the minimum detectable activity. Measurements were conducted using α/β discrimination, enabling the differentiation between alpha emitters (210Po) and beta emitters (210Pb and 210Bi) [12].

Results and discussion

Table 1 presents a comparison of various methodologies used to measure 210Pb, specifically within the context of tobacco. This comparison highlights the analytical aspects of the data collected. Each method described herein offers unique advantages and limitations. The choice of method depends on specific study requirements, including desired sensitivity, speed, sample type, and the availability of equipment at the research facility.

Table 1 Comparison of methods and preparation for the determination of 210Pb
Full size table

Gamma spectrometry with direct measurement of samples employs a low counting efficiency of about 4.25% and an extended counting time of 65,000–120,000 s [2,3,4]. Although this setup offers thorough analysis, its practicality is limited in rapid testing environments but is well-suited for detailed studies where maximum detection sensitivity is crucial. The extended counting duration is essential for achieving lower detection limits. The low-background gas flow proportional counter, utilizing precipitation techniques such as Ba (Ra, Pb) SO4, PbS, and PbCrO4, achieves a significantly higher counting efficiency of 35% and reduces the counting time to just 400 min [5]. It offers an efficient and swift analysis, suitable for scenarios that require rapid and sensitive results. However, it involves destructive sample preparation and the use of chemicals, leading to waste generation, thereby imposing certain operational limitations [5]. Alpha spectrometry, employing coprecipitation with lead sulfide, achieves a counting efficiency of 15% and a variable counting time of 400–80,000 min [5, 6, 8]. It should be noted that 210Pb is indirectly determined via 210Po. It features a low MDA of 0.3–4.5 mBq/g, highlighting its high precision and making it ideal for detecting very low levels of radionuclides in studies where maximum detection sensitivity is paramount [5, 6, 8]. A significant disadvantage is the extended period required to establish radioactive equilibrium when determining Pb through its daughter product, 210Po. Our method using liquid scintillation spectrometry achieves an exceptionally high counting efficiency of 82% and a remarkably short counting time of just 60 min. The sample preparation, including digestion, requires approximately 24 h, while the separation process is notably rapid, taking no more than 2 h. This method’s efficiency and speed greatly enhance its applicability in fast-paced research environments, providing a robust solution for the rapid and reliable measurement of 210Pb in diverse sample matrices. The measurement of 210Pb via the ingrowth of 210Bi can be accurately determined just 48 h post-separation, which is critical for the rapid development of analytical methods. Nevertheless, this method does use strong acids such as nitric acid and HCl, along with scintillation solutions, though waste volumes can be reduced by evaporation.

The detection limit for 210Pb using the Tri-Carb 3100 TR liquid scintillation spectrometer was established at 0.03 Bq, with a background pulse count of 7.3 cpm f (counts per minute), a measurement time of 60 min, and a separation efficiency of 0.82. The high detection efficiency offered by liquid scintillation spectrometry, compared to methods such as gamma spectrometry, underscores its superiority. For instance, the detection efficiency using a gamma detector is relatively low at only 4.25%, necessitating prolonged measurement times up to 90,000 s [5]. In contrast, our measurements, conducted using α/β discrimination, allow clear differentiation between alpha emitters (210Po) and beta emitters (210Pb and 210Bi). The bias between the reference value and experimental activity remained within 6.5% during the ingrowth period and showed no significant change after secular equilibrium was achieved. The fraction of 210Pb obtained through the AnaLig Sr01 procedure was repeatedly counted until equilibrium was achieved. The results are detailed in a previous publication [12].

Results from measurements conducted using the liquid scintillation detector for lead in tobacco solutions spiked with 210Pb, as determined by both the AnaLig Sr01 and Sr Resin methods, were analyzed (Table 2). The comparative analysis reveals that both Sr Resin and AnaLig Sr01 are effective methodologies for the determination of 210Pb. However, AnaLig Sr01 demonstrates slightly superior performance in terms of chemical recovery and reduced variability in measurement uncertainty. Both methods exhibit comparable detection efficiencies Sr Resin averaging 0.821 and AnaLig Sr01 averaging 0.823, which indicates similar technical effectiveness. Chemical recovery rates for Sr Resin vary widely, ranging from 76.5 to 89.8%, with an average of 85.2%. In contrast, AnaLig Sr01 demonstrates higher and more consistent recoveries, ranging from 82.3 to 100%, with an average of 94.4%. This consistency suggests that AnaLig Sr01 may be a more robust method under varying conditions.

Table 2 Results from measurements conducted using the liquid scintillation detector for lead in tobacco solutions spiked with 210Pb, as determined by both the AnaLig Sr01 and Sr Resin methods
Full size table

The activities measured by AnaLig Sr01 show a narrower range, from 248 to 283 mBq, with uncertainties tightly clustered around ± 13 to ± 14 mBq. The average activity for AnaLig Sr01 is calculated at 265 mBq with an uncertainty of ± 13 mBq. Compared to Sr Resin, AnaLig Sr01 presents a tighter distribution of activity measurements and a marginally lower average uncertainty, indicating more precise measurements. This trend toward consistency and precision, with a tighter spread in activity values and lower average uncertainty, may be indicative of better repeatability and potentially more accurate sample handling or analysis techniques.

We applied the methodology for determining 210Pb to real samples of tobacco and cigarettes available in Slovakia. The results were compiled into the article ʹDetermination of 210Pb Activities in Slovak Tobacco and Cigarettes: A Study on Radiological Risks and the Impact of Smoking,ʹ and submitted to a journal for review. For example, the weight of the tobacco that was analysed and separated onto 0.5 g of AnaLig Sr sorbent was 30 g. All measured values of 210Pb activity in tobacco and cigarettes exceeded the calculated MDA value, set at 5.12 mBq/g. For the application of a new methodology to various sample matrices, validation of the method is always necessary. The novelty of our article lies in describing and comparing the advantages of selected separation methods, AnaLig Sr01 and Sr resin, as well as comparing different methodologies for the measurement of 210Pb. Additionally, statistical comparisons of selected methods for a new matrix, such as tobacco, are described and presented.

Statistical testing of the determination of 210Pb with the AnaLig Sr01 method against the Sr Resin method

The validation process is crucial for identifying weaknesses or areas of improvement in a methodology, particularly when dealing with new matrices such as tobacco samples. By validating the method, we ensure its accuracy and effectiveness in quantifying lead content without significant interference from other substances. Validation also assesses the consistency of the methodology, examining whether consistent results can be obtained under the same conditions (repeatability). Through validation, essential performance characteristics like sensitivity, specificity, detection limit, quantitation limit, range, linearity, accuracy, precision, and robustness are determined. Understanding these parameters helps to define the strengths and limitations of each method. When comparing methodologies like Sr Resin and AnaLig, having validated methods ensures that the comparison is based on quantifiable, reliable data. This prevents misleading conclusions that could arise from comparing non-validated methods, where variable results may be due to methodological flaws rather than true differences. Ultimately, the validation process not only ensures the accuracy of our results but also drives method optimization, enhancing effectiveness and efficiency.

Regression diagnostics procedures were employed to examine the regression triplet and a paired t test was applied to compare the methods for the separation of 210Pb with AnaLig Sr01 [13,14,15,16]. Subsequently, the results obtained through both methods underwent statistical comparison. The results of regression diagnostics are presented in Table 3. The comparison employed a rigorous linear regression approach using the regression triplet (QC-EXPERT-Trilobyte program, Pardubice, Czechia). The linear regression model y = β0 + β1·x was utilized to test the following hypotheses: H0: β0 = 0, β1 = 1. The model critique indicated that the section β0 is equal to zero with a 17.1% statistical probability, which is greater than the chosen level of 5%. The interval estimation for β0 (− 103.1, 21.75) includes zero. The slope β1 is deemed equal to one, as its interval estimation (0.90, 1.37) encompasses unity. In the critique of the method, the coefficient of determination, i.e., the regression rabat, indicates that 94% of points correspond to a straight line, and the data fulfil a constant scatter with a probability of 97% (Fig. 1). The demonstrated linear model conclusively affirms that the residuals adhere to the Gaussian model without any autocorrelation and trend. Both analytical methods for determining 210Pb in tobacco-AnaLig Sr01 and Sr Resin produce consistent results.

Table 3 Building and testing of the regression model led to the straight-line model y = β0 + β1x, where y is the activity of 210Pb (mBq) determined with the AnaLig Sr01 method and x is the activity of 210Pb (mBq) determined with the Sr Resin method
Full size table
Fig. 1
figure 1

Scatter plot of the data set for AnaLig Sr01 versus Sr Resin tested for the determination of 210Pb in tobacco solution (using the regression triplet)

Full size image

The independent samples t test was employed to compare the means of two independent groups. Specifically, to determine whether the expected activity values of 210Pb with the AnaLig Sr01 method and Sr Resin method are different. The null hypothesis H0 in a two-sample t test assumes no significant difference between the means of two groups being compared. The t test confirmed the acceptance of the hypothesis regarding the determination of 210Pb activities with the AnaLig Sr01 method and the Sr Resin method in the tobacco solution (Table 3). The p-value (2.1) exceeds the significance level (0.536), leading to the retention of the null hypothesis (Table 4). This indicates that there is insufficient evidence to suggest a significant difference between the means. Variance equivalence test confirms that the activities are equivalent for both methods. The Kolmogorov–Smirnov Goodness of Fit Test (K-S test) compares the data with a known distribution and lets you know if they have the same distribution. Results confirmed that distributions are equal.

Table 4 The t test for determining 210Pb using AnaLig Sr01 versus extraction chromatography Sr Resin in the tobacco solution with added reference activity of 210Pb
Full size table

The verification of basic assumptions for the individual samples of AnaLig Sr01 method and Sr Resin method indicates that the data in all samples are independent and homogeneous without any outliers. The homogeneity test demonstrated a Gaussian distribution for individual samples. All dispersion tests confirmed agreement for the investigated AnaLig Sr01 sorbent, establishing identical dispersion.

Conclusion

The aim of our study was to statistically compare the suitability of two methods for determining 210Pb in tobacco: the molecular recognition sorbent method AnaLig Sr01 and the extraction chromatography sorbent method Sr Resin. The liquid scintillation spectrometer TriCarb 3100 TR was used for the measurement and determination of 210Pb activity in tobacco solution. Statistical analysis confirmed that the measured activities of 210Pb in the tobacco samples demonstrated good accuracy and precision for the described methods. The demonstrated linear model clearly proved that the residuals follow the Gaussian model without any autocorrelation and trend for all samples. The pair t test showed that all methods lead to consistently similar results. All analysed methods are fast and efficient for 210Pb separations in tobacco samples. The results confirmed the suitability of the 210Pb determination method in tobacco using both approaches. These methods are characterized by high detection efficiencies of 82% for both the AnaLig Sr01 and Sr Resin methods. Additionally, high yields were demonstrated, with 94% for AnaLig Sr01 and 85% for Sr Resin. These findings will be further explored in a forthcoming article, which will focus more broadly on tobacco and cigarette samples.