Patrice Fleury
COGEMA La Hague
France

Gilbert Deloge
COGEMA Marcoule
France

Martial Huver
EURISYS MESURES - COGEMA Group
Montigny le Bretonneux
France

The interpretation of non destructive waste assay results is not an easy task when a unique method of measurement is performed. A combination of measurement technics gives more information that it is not easy to interpret in line by non specialist operators. An interpreter software is the best way to give confident results from different assays. A description of the latest concept of interpretation of the different results delivered by the different measurement stations, developed for the conditioning centers of COGEMA La Hague and Marcoule sites, is presented in this poster.The final benefit was a significant reduction of volume of conditioned waste prepared for final storage.

Different measurement techniques are used for nondestructive waste assay. The most common are passive and active neutron counting, gamma spectroscopy, and in addition dose rate measurement.

Mainly depending on the radionuclide type to measure, the waste matrix density, and the waste matrix nature, each method has its own advantages and disadvantages.

To optimize results, several of these methods are often combined to complement each other, so as to increase the amount of information. On the other hand, it becomes difficult to correlate the results issued from the different methods, and to manage the uncertainties.

Interpreter is a software developed, first to satisfy the interpretation needs of the results issued from combined measurements, and second to manage the possible discrepancies between intermediate results. Interpreter carries out the synthesis in order to produce the declared radionuclide list, their alpha and beta activities, and their uncertainties.

In addition, Interpreter manage the stream of barrels to measure, depending on their class of emission : beta-gamma or alpha-gamma or alpha-neutron. Then, the waste packages can be sent to the appropriate assays according to their emission class.

The radionuclides present in the waste package can be :

• gamma emitters with energy and emission intensity allowing identification and quantization of the activities by spectrometry assay,
• pure beta or beta-gamma of low intensity allowing no identification or quantization by non destructive method,
• spontaneous neutron emitters (by alpha-n reaction or by spontaneous fission),
• induced neutrons emitters (by neutron activation).

The non destructive method of measurements processed on the radioactive components of the waste packages can be :

• gamma spectroscopy allowing :

• passive neutron assay giving :

• active neutron assay giving the fissile mass content.

Complementary information necessary to the interpretation of the results are provided by the predifined spectra concept.

This predifined spectra compiles the radiological characteristics of the waited elements to assay. The expected spectra is established by the producer by previous knowledge of the history of the waste (type of reactor, origin, process of treatment, date of production, etc...)

The predefined spectrum contents the following information :

• the Beta gamma composing normalized at 100% :
  each element of this composing is defined by

• the alpha composing normalized at 100% :
  each element of this composing is defined by

• the alpha-beta ratio

• the chemical form of the contaminant

The operation is based on a succession of steps of treatment :

• picking function before measurement

• interpretation after measurements

• selection of the declared gamma emitters measurable by gamma spectroscopy,
• calculation of the non measurable radionuclides by gamma spectroscopy,
• initiate a supplementary neutron measurement,
• verification of the coherence of the gamma spectroscopy results with the predefined spectra.

• Elaboration of a pertinent alpha spectrum in regard of the gamma spectroscopy results (with isotopic composition of Pu and U) and the predefined spectrum,
• elaboration of the alpha activity per radionuclide by combination of the different information collected :

The coherence of the results provided by the different measurement systems is based on the properties of the radionuclides:

• in passive mode, comparison of the 240 Pu equivalent mass measured by total neutron with the 240 Pu equivalent mass measured by coincidence neutron,
• in active mode, comparison of the 239 Pu equivalent mass measured by total neutron with the 239 Pu equivalent mass measured by coincidence neutron,
• comparison of the alpha activity calculated from passive neutron assay with the alpha activity derived from the gamma spectroscopy,
• comparison of the alpha activity calculated from active neutron assay with the alpha activity derived from the gamma spectroscopy,
• comparison of the ratio Pu even isotopes/odd isotopes calculated by neutron assay with the ratio calculated from the gamma spectroscopy results.

The main results delivered by the different assays are compared to the expected spectrum :

• isotopic composition of Pu with the isotopic composition of the expected spectrum,
• isotopic composition of U with the isotopic composition of the expected spectrum,
• beta-gamma measured and calculated spectrum with the expected spectrum,
• alpha-gamma measured and calculated spectrum with the expected spectrum,
• alpha/beta ratio with the expected spectrum,
• Pu even isotopes/odd isotopes ratio with the ratio of the expected spectrum,
• 244 Cm/ 240 Pu ratio with the ratio of the expected spectrum,
• 235 U/239 Pu ratio with the ratio of the expected spectrum.

The different comparisons are led by coherence laws which lead to the most realistic results delivered in the final report. In case of too large discrepancies, attention is sent for detailed analysis. the edition of the sources of discrepancies are indicated in order to help the specialist analysis.

The comparisons are based on dedicated rules based on the radioactive laws in conjunction with the chemical and physical laws. These laws need the edition of the parameters needed for the comparisons. in order to give the base of the parameters and the values for comparisons and calculations, a nuclear data table is accessible to the competent operators

The flow chart of Figure 2 gives the general process of orientation to the different measurement stations based on a expected spectra and post-coherence analysis.

Some examples of coherence tests:

• on pure Uranium alpha waste , detection of Pu total absorption peaks : isolation of the drum,
• on pure Plutonium alpha waste, detection of 235 U total absorption peak : isolation of the drum,
• on pure beta waste, detection of U or Pu total absorption peaks : supplementary measurement in the alpha gamma spectroscopy station,
• uncoherence between the Pu mass calculated by spectroscopy and by neutron coincident analysis : maximum value taken in account and in some cases operator analysis,
• uncoherence between the Pu mass calculated by total neutron counting (with the chemical form knowledge) and the neutron coincident counting : maximum value taken in account and in some cases operator analysis,
• uncoherence between the expected isotopic composition of Plutonium and the measured isotopic composition by gamma spectroscopy : research of incoherence between the alpha gamma spectroscopy station spectrum and the beta gamma spectroscopy station spectrum, research of significant beta activity increasing the detection limit and if yes supplementary measurement in the neutron analysis station.
• different results of activities calculated by two different assays with recovery of the uncertainies : the activity is declared on the mean moderated value.

The interpretation concept has led to a more automated and more precise installation of waste barrel conditioning and sorting. Since 1996, date of the improvement of the workshop of La Hague, 80% of the drums measured had been declared proper for a steel conditioning, whereas without interpretation, they were declared for fiber concrete overpack. A volume reduction of 3 has been the direct benefit of both improvement of the measurement by combining different method of assays under the interpretation software control.

BACK