2.6.16 - Multiple Radionuclide Analysis

Usually, environmental monitoring or biological samples to be measured by LSC contain more than one radionuclide. A mixture of β-emitting radionuclides produces a continuous spectrum from almost zero energy to E_max; therefore, all overlap to some degree. Chemical separation procedures before its measurement are often tedious and time consuming. If only two or maximum three β-nuclides with sufficient differences in maximum energies are present, then the exclusion method described in chapter 2.1.2. is a suitable solution. The radionuclide with the highest energy will have a limited part of the pulse height spectrum free of any overlap. The interpolated contribution to the entire spectrum can thus be subtracted and the procedure repeated multifold.

Due to the technological advances of computer aided spectra processing and development of new mathematical approaches for decoding complex spectra, LS spectrometry has become an excellent methodology for estimating the radionuclide composition of a multinuclide sample. Spectrum deconvolution methods such as unfolding, stripping or peak fitting, or applying interactive spectrum stripping software using digital spectrum libraries (“digital overlay technique” DOT [Rundt 1992]) allow the quantification of individual radionuclides in a multi-nuclide spectrum.

The main challenge of the method is the correct splitting of the multicomponents of the spectrum into individual initial components with the aid of computerized software. The method is based on setting as much counting regions as radionuclides are present in such a way, that there are similar spillup and spilldown of pulses from the b-emitters present. After defining the counting regions, quench correction curves for each radionuclide in each counting region have to be prepared.

The application of multi-channel analyzers and the direct computerized data processing has enabled analyzing numerous α- and β-emitting radionuclides simultaneously in a radionuclide mixture including γ- and X-ray emitting radionuclides as well as EC nuclides with low energy Auger electrons [Malinovsky 2002], [Ermakov 2006], [Altzitzoglou 2008], [Nebelung 2009], [Verrezen and Hurtgen 2000] and more recently [Lee et al. 2017]. A combination of spectral unfolding by LSC and γ-measurement by NaI(Tl) spectrometer has been applied for waste solutions [Takiue 1999]. For a more detailed survey of multinuclide analysis see [L’Annunziata 2012].

RadSpectraDec presents a commercially available software package that allows the decoding of complex spectra [Ermakov 2006]. It is based on the superposition of individual reference spectra, taken from a previously created radionuclide library of spectra, which allows obtaining information on the sample composition.

As an example, we describe below a multiple window deconvolution methodology with internal standardization and spectrum deconvolution for measuring low-energy β-emitters in multi-labeled samples, containing high-energy β-impurities. The method does not require setting up quench correction curves and exact values of reference activity [Verrezen and Hurtgen 2000].

Materials and Equipment

• Aqueous standard solutions of radionuclides involved

• Carbon tetrachloride (CCl₄) or chloroform (CHCl₃) for quenching

• Gelating cocktail (e.g. OptiPhase HiSafe III)

• Low level water sample (reversed osmosis water)

Measurimg Procedure

For each radionuclide known to be present in the sample (impurity and nuclide of interest), a set of counting vials is prepared as is described below:

1. Background vial: 10 mL of reversed-osmosis water and 10 mL gelating cocktail

2. Sample vial: 1 mL sample, 9 mL of reversed-osmosis water and 10 mL of gelating cocktail

3. Reference vials: 1 mL sample, 1 mL of an aqueous reference solution of appropriate radionuclides present in the sample, 8 mL of reversed-osmosis water and 10 mL gelating cocktail

4. All vials are counted (2 h thermal equilibration period before measurement) and β-spectra are recorded.

5. The influence of quenching is evaluated by adding appropriated amounts of CHCl₃ to the scintillator cocktail before it is added to the measuring vials. The same quenching parameter is used for characterization of all the samples.

Data treatment

1. Calculate the net values of the individual spectral data

2. Correct the spectra from background contribution

3. Reconstruct the spectrum of the contaminant in the lower energy range

4. Restore the net spectrum of the nuclide of interest (in the lower energy range) by substracting the calculated net spectrum of the impurity from the net sample spectrum

Evaluation

The resulting spectrum of a sample containing a β-emitting radionuclide, contaminated with one or more β-emitting nuclides, corresponds to the sum of the individual spectral contributions generated by the background, the nuclide of interest and the impurities.

The determination of low energy β-activities in the presence of high energy β-impurities is possible by means of deconvolution algorithms; for more details see literature [Verrezen and Hurtgen 2000].

Altzitzoglou T. 2008: Radioactivity determination of individual radionuclides in a mixture by liquid scintillation spectra deconvolution; Appl. Radiat. Isot. 66 (2008) 1055-1061

L’Annunziata M.F. 2012: “Handbook of Radioactivity Analysis”, Chapter 15, 3^rd Edition 2012, Elsevier

Ermakov A.I., Malinovsky S.V., Kashirin I.A., Tikhomirov V.A. and Sobolev A. 2006: Rapid analysis of radionuclide composition (screening) of liquid samples via deconvolution of their LS spectra; in: S. Chalupnik et al. “LSC2005 Advances in Liquid Scintillation Spectrometry”, pp 89-98, Radiocarbon 2006, Tucson

Lee U., Bae J.W. and Kim H.R. 2017: Multiple beta spectrum analysis based on spectrum fitting; J. Radioanal. Nucl. Chem. (2017) 617-622

Malinovsky S.V., Kashirin I.A., Ermakov A.I., Tikhomirov A.I., Sobolev A.I., Kaihola L.L. and Fedotov A.A. 2002: New software for analyzing complex spectra obtained with ultra low level liquid scintillation spectrometer ‘Quantulus’; in: S. Möbius et al. “LSC2001 Advances in Liquid Scintillation Spectrometry”, pp 119-126, Radiocarbon 2002, Tucson

Nebelung C., Jähnigen C. and Bernhard G. 2009: Simultaneous determination of beta nuclides by liquid scintillation spectrometry; in: J. Eikenberg et al. “LSC 2008 Advances in Liquid Scintillation Spectrometry”, pp. 193–201, Radiocarbon 2009, Tucson

Rundt K. 1992: Digital overlay technique in liquid scintillation counting–a review; Radioact. Radiochem. 3 (1992) 14-25

Takiue M., Natake T. and Fujii H. 1999: A hybrid radioassay technique for multiple beta-emitter mixture using liquid and NaI(Tl) scintillation spectrometers; Appl. Radiat. Isot. 51 (1999) 429-434

Verrezen F. and Hurtgen C. 2000: A multiple window deconvolution technique for measuring low-energy beta activity in samples contaminated with high-energy beta impurities using liquid scintillation spectrometry; Appl. Radiat. Isot. 53 (2000) 289-296