Determination by x-ray fluorescence of Caesium and other elements from Hanford waste simulates in the application of Synroc for high level radioactive waste remediation

Abstract

At the Hanford Reservation near Richland, USA, there is a complex of 149 single-shelled tanks (SSTs) and 28 double-shelled tanks (DSTs) storing 9.7 x 1018 Bq (260 MCi) of radioactive waste awaiting remediation [1]. The Synroc process [2] developed by the Australian Nuclear Science and Technology Organisation (ANSTO) at Lucas Heights may be able to provide an effective solution for permanent containment and immobilisation of the highly radioactive caesium (137Cs) component of the Hanford waste. Decontamination of the liquid component of the waste, plus sludge washings, involves the absorption of caesium (Cs) on a suitable inorganic ion-exchange material. For Synroc, Cs-loaded sodium silico-titanate [4] would be a suitable precursor. We are preparing this ion-exchanger and a screening study using distribution coefficients (Kd) as a measure of efficiency is being undertaken. This work entails precise analysis of Cs and this forms the subject of the present paper.

The concentration of caesium in Hanford waste simulates can be determined by various techniques. The Los Alamos National Laboratory has used radioactive tracers for simulates with a caesium concentration of 6 μg.l-1 [3]. Sandia National Laboratories have used flame atomic absorption spectrophotometry (AAS) [4]. At Lucas Heights Research Laboratories Cs in solution has been determined by electrothermal atomic absorption spectrophotometry (AASET) which has a detection limit of 2 μg.l-1 [5].

When Cs is absorbed on a solid matrix, analysis is further complicated. Quantitative chemical analysis by solution techniques requires sample preparation. An initial step of taking up minerals in a suitable flux (e.g. lithium metaborate) is uncomplicated, but with compounds containing titanium (Ti), the next step of making a solution, requires a special procedure involving a quenching liquid containing 5 % (v/v) of 30 % (w/v) hydrogen peroxide mixed with 2.5 % (w/v) tartaric acid dissolved in 3 % (v/v) nitric acid [6]. To avoid problems associated with having to quantitatively dissolve titanates to form stable solutions for subsequent analysis of Cs and other elements by atomic absorption spectrophotometry or inductively coupled plasma mass spectrometry (ICP-MS), analysis of solid material by wavelength dispersive X-ray fluorescence (WDXRF) was selected for this study.

An advantage of WDXRF is that it is a multi-element technique capable of determining Cs and all other elements present in the liquid Hanford simulate and solid ion-exchange materials. Ti analysis by WDXRF is a common standard procedure for rocks and minerals and the analytical parameters are well known. XRF has also been used for the analysis of the chemical composition of sodium titanate [7] so the analysis of sodium silico-titanates should have few complications. For XRF the form of sample preferred for presentation to the detector is a glass disc. The major concern with Cs analysis that there may be some loss of this element during the preparation of the glass disc, because Cs had been found to volatilise from borosilicate glass melts [8]. While powders can be analysed by XRF, relatively large amounts ( 6 to 10 g ) are needed and there are other technical problems such as the difficulty of duplicating particle size and controlling the homogeneity of the sample, for the reasons analysis in the form of powders is not suitable for precision analysis.