TECHNICAL NOTE - Semi-Continuous On-Line Uranium Ore Analyzer

Introduction The efficient process control of a uranium mill and its associated mining operation requires a nearly continuous knowledge of the uranium concentration in the ore. Generally, the approach is to use laboratory assays of grab samples from the mill feed belt. In some cases, elaborate and expensive systems have been used to ensure random sampling, but even with these systems, mass-balance discrepancies still exist. There is a requirement for an on-line instrument that can give a prompt, accurate analysis of a large portion of the feed stock. The authors have recently evaluated a laboratory system that achieves these goals using neutron activation and delayed neutron counting. The development of the on-line uranium ore analyzer is a consequence of previous work done at the Whiteshell Nuclear Research Establishment (WNRE) (Campbell et al., 1978 and 1981), and is based on the emission of delayed neutrons after the neutron irradiation of fissionable materials. The mechanism of delayed neutron emission has been described (Keepin, 1965), but briefly is as follows. The fission fragments resulting from irradiation are in an excited state. Certain of these delayed fission products, precursors, decay to a more stable state by the emission of a delayed neutron. The delayed neutrons can be divided into six groups with effective half-lives ranging from 0.2-55 sec. It is important to note that more than 50% of the delayed neutrons are emitted within the first 6 sec after irradiation; this has an influence on the design of the analysis instrument. Measurement Cycle The measurement cycle involves a neutron irradiation, a delay time, and a counting period. The laboratory demonstration model of the ore analyzer comprises a californium-252 neutron source, a source chamber, a pneumatic transfer facility, and a neutron irradiation and counting module (Fig. 1). The ore sample is placed in the irradiation chamber. The neutron source is transferred to the center of the chamber for a 5-sec irradiation while the helium-3 thermal neutron detectors are switched off. As the source returns to the storage chamber, the detectors are switched on and the delayed neutrons counted. The total assay cycle is 16 sec. The current instrument design calls for the ore to be cut from the main belt every 16 sec by a mechanical arm and dropped into the analyzer directly, or via another belt. The cycle starts when there is a sufficient mass of ore in the irradiation chamber to depress a switch and stops when the delayed neutron count is finished. The ore is then dropped through the bottom of the analyzer chamber and returned to the main belt. The counts from each cycle are accumulated and the data reduced to give the desired information. Key for Fig. 1 A-Helium-3 neutron detector B-Wax moderator for neutrons C-Irradiation chamber D-Funnel for receiving ore E-Gate F-Transducers G-Biological shield for neutron source H-Transfer tube for neutron source I-Neutron source J- Microprocessor K-Accumulator for hydraulics L-Control panel M-Electro-hydraulic servo valve to control movement of the source N-Electro-hydraulic proportional valve to control movement of the gate O-Hydraulic power unit Irradiation and Counting Equipment Irradiation of the ore is achieved using a 1 X 10 7 neutron*s-1 californium-252 neutron source. The source is contained in an iron capsule and pneumatically transferred between the irradiation/counting and storage chambers. The irradiation/counting chamber possesses 12, 600-mm (24-in.) long helium-3 thermal neutron detectors embedded in an annulus of wax (Fig. 1). Two sets of six detectors are each interfaced to a preamplifier, amplifier and single-channel pulseheight analyzer system, and the two systems are connected to a scaler. A special timing unit controls the transfer of the source and the on and off switching of the detector assembly, in sequence with the movement of the source. Results and Discussion In the experimental demonstration, using the 107 neutron.s-1 source, a 10% uranium ore was measured to ± 10% (2o total error) in 10 cycles and ±6% (2o) in 100 cycles. Since a larger source was not available, it can be statistically predicted that a 109 neutron-s-1 source could be used to measure a 0.1% ore to ± 10% (20) in 10 cycles. The uranium can occur as discrete nonhomogeneous particles in the ore, and the error caused by such a distribution of particles is ± 4% (2o), as determined by