Method and Apparatus for Magnetically Determining the Gd.sub.2 O.sub.3 Content in UO.sub.2 Fuel Pellets While Eliminating the Effect of Ferromagnetic Impurities

This invention relates to methods and apparatus for determining the concentration of burnable poisons in reactor fuel pellets. More specifically, this invention relates to a method and apparatus for measuring the Gd.sub.2 O.sub.3 content in UO.sub.2 fuel pellets wherein ferromagnetic impurity moments are saturated with a dc bias magnetic field.

Uranium oxide fuel pellets in nuclear reactors will contain gadolinia (Gd.sub.2 O.sub.3) as a burnable poison in increments of 0.5 w/o from 0.0 w/o to 8.0 w/o. The pellets are assembled into fuel rods and are clad, for example, with zircalloy. Each completed fuel rod may contain up to six zones with different gadolinia concentrations. A method is required to determine that the correct concentration and distribution of gadolinia is present within a completed fuel rod assembly.

Magnetically UO.sub.2 and Gd.sub.2 O.sub.3 are paramagnetic and have susceptibilities, X, of 8.74 .times. 10.sup.-6 emu/g-Oe and 147 .times. 10.sup.-6 emu/g-Oe, respectively. The processing of fuel pellets for use in reactors typically introduces up to 500 ppm of elemental iron and/or ferromagnetic alloys as impurities.

The magnetic susceptibility dM/dH of uranium oxide fuel pellets has been found to increase with the addition of gadolinia. Ferromagnetic inclusions in fuel pellets complicate the gadolinia determination by adding susceptibilities proportional to the ferromagnetic content. The susceptibility of ferromagnetic inclusions decreases, however, at high applied fields.

The gadolinia content of fuel pellets may be nondestructively determined by measuring the alternating current susceptibility using an inductive technique. Ferromagnetic inclusions are saturated with a high, direct current magnetic bias field which is applied during the susceptibility measurements.

It is, therefore, an object of this invention to provide nondestructive methods for determining the gadolinia concentration and distribution in uranium oxide fuel rods.

Another object of this invention is to minimize the measurement errors which occur when the chemical content of materials with ferromagnetic inclusions is determined from magnetic susceptibility measurements.

The novel features believed characteristic of the present invention are set forth in the appended claims. The invention itself, together with further objects and advantages thereof, may best be understood by the following detailed description, taken in connection with the appended drawings in which:

FIG. 1 is a magnetization curve of a fuel pellet;

FIG. 2 is an apparatus for determining the gadolinia content in uranium oxide fuel rods; and

FIG. 3 is a plot of output signal vs. distance along a fuel rod, measured in accordance with the invention.

The magnetic susceptibility of UO.sub.2 is 8.74 .times. 10.sup.-6 emu/g-Oe and that of Gd.sub.2 O.sub.3 is 146.8 .times. 10.sup.-6 emu/g-Oe. The susceptibility of UO.sub.2 reactor fuel pellets will, therefore, increase for each 0.5 w/o addition of Gd.sub.2 O.sub.3 in the UO.sub.2, as (U,Gd)O.sub.2. Ferromagnetic inclusions in the fuel pellets complicate the gadolinia determination by adding susceptibilities proportional to the ferromagnetic content. Fortunately, the susceptibility of ferromagnetic inclusions decreases dramatically at high applied fields. The solid curve of FIG. 1 shows an M/H curve of a fuel pellet containing 0.5 w/o Gd.sub.2 O.sub.3 and ferromagnetic impurities (magnetically equivalent to 25 ppm iron). The dashed curve of FIG. 1 indicates the slope of the curve of a fuel pellet without iron impurities. The problem of ferromagnetic inclusions is, therefore, minimized by measuring the susceptibility of fuel pellets in a magnetic field which is large enough to saturate the ferromagnetic component. The saturating field may be produced by biasing the fuel pellet with a dc magnetic field component. It is evident from FIG. 1 that the magnitude of the bias field should be at least 4 kOe. In practice, a value of 5 kOe is satisfactory, but a larger value (e.g., 8 kOe) yields somewhat improved resolution.

FIG. 2 is apparatus for measuring the ac susceptibility of a fuel pellet using an inductive technique with a coaxial dc bias field. Techniques for inductive susceptibility measurements are described, for example, in Chapter 8 of Magnetism and Mettalurgy, edited by A. Berkowitz and E. Kneller, Academic Press, New York, 1969.

A pair of Helmholtz coils 10 and 11 are driven from an audio power amplifier 12 to produce a homogeneous ac magnetic field. Two coils 13 and 14, with substantially identical area.turns (NA) are placed between the Helmholtz coils 10 and 11 in the ac magnetic field. The coils 13 and 14 are connected in series opposition so that the voltage induced by the ac field approximately cancels. When a magnetic specimen is placed near one of the coils, an unbalanced voltage is produced which is proportional to the susceptibility of the sample dM/dH. The output voltage from the series opposed pickup coils 13 and 14 is fed to the input of a phase-lock amplifier 15. The output of the amplifier drives a recorder 16. The reference output signal from the phase-lock amplifier 15 is applied to the input of the audio power amplifier 12 to provide drive for the Helmholtz coils 10 and 11.

Four fuel pellets 17, 18, 19, and 20 were stacked in a zircalloy tube 21 which was translated along the axis of the pickup coil 14 during a measurement. The poles of a dc electromagnet 22 and 23 were aligned coaxially with the Helmholtz coils 10 and 11 and the pickup coils 13 and 14 to provide a dc bias field. This configuration gives maximum sample-to-coil coupling. Other coil and magnet configurations may also be used with a reduction in coupling efficiency.

The operating frequency was 82 Hz. The frequency is not critical but should be chosen to minimize mechanical resonances in the sample and measurement system and should be low enough to minimize skin effects and thus assure that the magnetic field will penetrate a nonmagnetic, conductive shield: that is, the zircalloy tubing 21. For zircalloy tubing with a resistivity of 7 .times. 10.sup.-7 ohm-meters, the penetration depth .delta. at 80 Hz is approximately one-half centimeter.

By way of example, in typical experimental apparatus, each Helmholtz coil is approximately thirteen centimeters in diameter and comprises 525 turns of No. 30 Formex.sup.R coated wire. Fields produced were approximately 15 Oe at the operating frequency. The pickuo coils were approximately 1.4 centimeters inside diameter, approximately 0.25 centimeters high and included 2000 turns of wire. Four fuel pellets were measured containing, respectively, 0, 0.5, 2.5, and 4.0 weight percent Gd.sub.2 O.sub.3 in UO.sub.2. The pellets were approximately 1.06 centimeters in diameter and 1.08 centimeters long.

FIG. 3 is a curve of output signals from the phase-lock amplifier vs. position of the sample tube 21 for a tube containing the four fuel pellets described above. The gadolinia content of the fuel pellets and their relative position in the tube may be readily discerned.

Alternately, the dc bias magnetic field may be provided by an axial field permanent magnet disposed around one of the pickup coils.

The methods and apparatus of the present invention allow nondestructive determination of the gadolinia content and distribution in fuel rod assemblies. A dc bias field saturates ferromagnetic inclusions and eliminates errors which might, otherwise, interfere with measurement accuracy.

While the invention has been described in detail herein in accord with certain preferred embodiments thereof, many modifications and changes therein may be effected by those skilled in the art. Accordingly, it is intended by the appended claims to cover all such modifications and changes as fall within the spirit and scope of the invention.