Laboratory application of NaI(Tl) γ-ray spectrometry to studies of natural radioactivity in geophysics

https://doi.org/10.1016/S0969-8043(00)00123-8Get rights and content

Abstract

We implemented a laboratory apparatus based on a NaI(Tl) γ-ray spectrometer for the assessment of naturally occurring radioactive materials. This apparatus is devoted to the quantitative determination of U, Th and K in rock and soil samples. We analysed the calibration procedures with particular emphasis to factors which affect accuracy and detection limits. A first series of measurements were compared to those obtained by means of high energy γ-ray spectrometry. A discussion was made on factors restricting the approach based on scintillation spectrometry, such as secular equilibrium and radon encapsulation.

Introduction

The γ-ray spectrometry method for the determination of naturally occurring radioactive materials (NORM) is of great interest in Earth’s sciences. There is a number of possible applications spanning from ore exploration to environmental radiation monitoring problems, most of them involving the determination of the U, Th and K amount in soils and rocks. These elements may be used as tracers also in non-radioactive processes producing paragenesis associated with NORMs (Subjender Reddy and Venkat Rao, 1995). In geophysical studies, the heat produced by radioactive decay in rocks is of fundamental importance in understanding the thermal history of the Earth and interpreting the continental heat-flux data. To this regard, 40K, 235U, 238U and 232Th are the most important radioactive nuclides. Such heat sources contribute significantly to the heat flowing out from the Earth, since they are abundant, their heat production is sufficiently large and they have been effective during a time comparable with the age of the Earth (Rybach, 1988, Verdoya et al., 1998).

Van Schmus (1995) recently gave a compilation of the abundance of NORMs in terrestrial materials. As one of the alkali metals (Li, Na, K, Rb, and Cs), with an ionic radius of about 1.6 Å, potassium is one of the so-called large ion lithofile (LIL) elements. The isotopic abundance of 40K, in weight, is 0.01193%. Potassium tends to get concentrated in crystal rocks, such as granites and shales, and in general is a primary constituent of several common rock-forming minerals.

Most of natural uranium is formed by 238U (99.28%) and 235U (0.71%). Natural thorium is, instead, totally constituted by 232Th. Also uranium and thorium are LIL elements, but they are found only in accessory minerals in common rocks or as minerals in primarily or secondary in deposits. Therefore, their overall abundance is much lower.

The main component of magmatic rocks like feldspar, quartz, olivine hornblende, pyroxene and mica are practically free of uranium and thorium. The accessory minerals, such as zircon, rutile, titanite and others have greater atom radii in their original lattice, and uranium and thorium can be substituted for the original atoms. During the process of magma solidification the main portion of U and Th is concentrated in the remaining, more acid portion of the liquid magma, which finally solidifies as granite. The particular temperature–pressure history during the solidification is decisive for the element content (Chiozzi et al., 1998).

This paper deals with the recent implementation of a NaI(Tl) γ-ray spectrometer at the Geothermal Laboratory of the University of Genova for the assessment of NORMs of relevant interest in terrestrial heat-flux studies. The equipment was calibrated taking into account the possible interference of each nuclides. A series of quantitative determination of 40K, 238U and 232Th in several types of rocks was then carried out. The results were compared with those obtained by means of high resolution γ-ray spectrometry.

Section snippets

Measuring system

Our γ-ray scintillation spectrometry system is an EG & G-Ortec NaI(Tl) detector (7.62 × 7.62 cm), connected to a multichannel analyser PC board (2048 channels) equipped with a spectrum stabiliser for automatic gain shift compensation. The detector is surrounded by a 5 cm thick led shield which smoothes the background γ-radiation. Polyethylene Marinelli beakers were used both for standards and analysed materials. A volume of 500 ml was adopted with sample mass ranging approximately from 0.650 to

Discussion

Secular equilibrium between 226Ra and 222Rn (238U series) can be practically assumed after about five to seven times the half-life (3.82 d) of the daughter 222Rn. Aware of the fact that polyethylene is permeable to radon, we monitored the total spectrum count rate versus time for five volcanic rocks (basaltic-andesite, andesite and rhyolite). The first count was performed just after sealing the Marinelli beaker and the different samples were measured at the same days. The relative total count

Conclusions

Although the use of generalised calibration procedure for NaI(Tl) γ-ray spectrometry is in principle preferable, our results show that a simplified approach can be adopted, provided that almost pure K, U and Th standards are available. The equipment implement in our laboratory offers good detection limits and acceptable uncertainty in activity determinations allowing its use in geophysical applications such as the determinations of concentration of heat production elements in rocks and soils.

A

References (10)

There are more references available in the full text version of this article.

Cited by (83)

  • Study of water radioactivity transfer from telluric origin in the Amber Mountain, Antsiranana, Madagascar

    2021, Scientific African
    Citation Excerpt :

    The soil samples were counted for 12 h and water samples for 24 h. Measurements of known masses of samples enabled determination of the specific activity of 40K, 238U series and 232Th series. The method described by Rybach was used to process the spectra obtained [14,13]. This method targets three regions of interest corresponding to three isolated intense total absorption peaks of 40K (1461 keV), 214Bi (1764.5 keV) and 208Tl (2614.5 keV), as shown in Table 1.

View all citing articles on Scopus
View full text