A New Version of MGA for Highly Attenuated Pu Samples

Introduction

Fast and accurate determination of the uranium enrichment and/or the isotopic composition of plutonium is essential in nuclear disarmament and stockpile stewardship. This includes the traditional verification and confirmatory measurements of Special Nuclear Materials (SNM) like those performed by Euratom and the IAEA, as well as Material Protection, Control and Accountability (MPC&A) applications for SNM identification at the nuclear facilities by their operators. High resolution gamma spectrometry coupled with the Multi Group Analysis (MGA) code (Gunnink 1987, 1990) has been used successfully for this purpose by both Euratom and the IAEA for years. The performance of its various versions has been tested extensively (D. D’Adamo et al. 1990, S. Abousahl et al. 1996). The results are generally quite accurate regardless of sample age, geometry, chemical composition or attenuation between the sample and detector.

However, older versions of MGA often required skilled and highly qualified personnel to correctly set up the instrument and to interpret the results and messages. Together with Dr. Gunnink, CANBERRA has enhanced the MGA code to be more robust for situations where the counting statistics are far from ideal (Verplancke et al. 1995). In addition, the hardware typically used for field plutonium measurements has been improved to provide better resolution and stability (Koskelo at al. 1997).

Recently, materials have been encountered that are stored in lead lined containers that greatly attenuate all low energy radiations. Although previous versions of MGA allowed the user to include data from the high-energy region (200 - 1020 keV) in the analysis process, they required the use of two separate detectors since the low energy region was still the primary region of analysis. Therefore, a new analysis mode has recently been added to MGA to allow a complete analysis to be made based on only the high-energy region of a spectrum. In addition to this highenergy region only mode, the newest version of MGA can still analyze the plutonium abundances in the traditional single low energy spectrum, and the combination of a low energy and a high energy spectrum modes. But it is the new high-energy only analysis mode that allows for the measurement and analysis of highly attenuated samples. In this paper, we will compare the results from low energy only and high-energy only measurements and show that good results can be obtained with this new analysis mode.

Experimental 

The facilities at Institute of Physics and Power Engineering (IPPE) have different plutonium samples for test purposes. These include plutonium metal (with 239Pu content of 88% and 95% ) and some special samples of PuO2 with 239Pu content of about 67%, 77% and 87%. The plutonium metal samples are in the form of disks in stainless steel cladding of 47 mm diameter and 3.16 mm in height. The plutonium oxide sample is contained in a 6 mm diameter tube and covered by stainless steel foil. The main characteristics of the samples used for the measurements in this study are shown in Table 1.

From each sample type we chose an item for measurement and analyzed it for its isotopic composition and compared the results with the known, so called “passport” values. The “passport” data of the plutonium samples is shown in Table 2. The declaration dates for the samples 981, 1124 and P37 are June, 1973, July, 1969, and February 1, 1988, respectively. 

Two types of High Purity Germanium detectors were used, a planar detector, CANBERRA model GL0510R with a resolution (FWHM) at 122 keV of 550 eV, and a coaxial detector, CANBERRA model GC1818 with resolution of 680 eV at 122 keV and 1.73 keV at 1332 keV. Low energy spectra were collected with the planar detector and the high-energy spectra with the coaxial detector. A CANBERRA Inspector MCA was used to collect the data and provide the signal processing for both detectors. For the “classic” low energy analysis with MGA we used a gain setting of 0.075 keV/channel, and a shaping time of 2 µs. The dead time was 15-20% depending on the sample and the presence of lead shielding. For the high-energy only analysis we used a gain setting of 0.250 keV/channel, and kept the shaping time at 2 µs. The dead time for these measurements was 15-35%.

Each sample was measured with and without a thin lead shield between the sample and the detector. For the unshielded measurements we used two or three 0.3 mm Sn plates during the measurements of the plutonium disks to reduce the effect of the 59 keV 241Am gamma radiation. For the shielded measurements, the thickness of lead plates was selected to reduce the intensity of all gamma rays below 210 keV. For samples 981 and 1124, we used 3 mm of lead, for sample P37 we used 4 mm of lead. Three series of experiments were performed for each sample for both the shielded and unshielded measurements: spectra with poor statistics, with average statistics, and with good statistics. This was achieved by varying the live time from sample to sample from a few minutes to about 60 minutes.

Results and Discussion 

The measurement results for the non-shielded samples for both the “classic” low energy MGA analysis and the new high-energy MGA for test sample 981 are shown in Table 3. 

The top part of the table shows the results for measurements with a short count times and the bottom part the results for measurements with long count times. The known “passport” values are shown in the second column for both portions of the table. The column marked “Classic” MGA shows the results for the normal low energy MGA analysis followed by the percent difference relative to the “passport” values. The last two columns show the results obtained with the new high-energy MGA mode and their percent difference from the “passport” values. Note that we have purposely excluded the 242Pu results from these tables. It cannot be measured directly, and a comparison of its reported values to the “passport” values is not a measure of how well MGA can analyze the spectrum but rather a measure of how well the correlation equation happens to match these particular test samples. In addition, each MGA result column represents results from a single measurement from the multiple measurements that were made for each sample.

Overall, the results are quite good for the main isotopes and there is not much difference between the poor statistics and the good statistics. The uncertainties reported by the two modes of MGA are not shown in the table. However, a statistical analysis indicates that they are appropriate for the observed deviations from the “passport” values. The results for sample 1124 show very little difference the poor statistics and the good statistics, and have not been included here. The results for test sample P37 are shown in Table 4. For these results, the difference is a little bit more noticeable, particularly in the high-energy analysis.

The “classic” MGA analysis cannot be performed on the measurements with the lead-shielded containers. All other techniques that use low-energy regions have the same problem. The lead lining prevents the low energy gamma rays that they use as a basis of the analysis from reaching the detector. However, the new high-energy only MGA analysis has no problems with the lead lined samples as long as the gamma rays above 300 keV can be detected and measured. The results with the high-energy analysis mode for test samples 981, 1124 and P37 encased in lead are shown in Tables 5, 6 and 7, respectively. Note that each of the MGA results columns again represents only one measurement of the multiple measurements that were made with each sample. 

It can easily be seen from the tables that there is generally an improvement in the analysis results as a function of the statistics. But it should be noted that even the good statistics spectra were all measured for less than 30 minutes. Again, the reported uncertainties from the MGA analysis are not shown on these tables for the sake of clarity. However, just like in the case of the unshielded sample measurements, a statistical evaluation of the replicate measurement results and a comparison of the reported uncertainty against the deviation from the “passport” values shows that the uncertainties reported by MGA are appropriate. 

Note that the declared date for two of the samples is 25 to 30 years ago. This means that the ²⁴¹Pu has gone through nearly two half-lives of decay. The comparisons made here were done simply by decaying the measured results back to the declared date for convenience, since the MGA code easily accommodates such a calculation. However, with such old material, most of the ²⁴¹Pu has decayed to ²⁴¹Am. This presents a problem in the 600 keV region which is crucial for the measurement of ²³⁸Pu and ²⁴⁰Pu in the high-energy mode. The problem is that the 662 keV peak of ²⁴¹Am, and it's Compton continuum, begin to mask the region that must be analyzed. However, this is a problem that applies to all codes using the high-energy region - not just to MGA.

Conclusions

The measurement results for the unshielded samples show that both modes of MGA are capable of analyzing the data effectively. Due to the spectral characteristics, the high-energy analysis generally exhibits somewhat larger deviations from the “passport” values than the low energy analysis for the same amount of count time. At the same time these larger deviations are accompanied by larger error bars for the reported values. A statistical evaluation of the uncertainties of replicate measurements shows that the uncertainties reported by MGA in either mode are appropriate relative to the deviations from the “passport” values. It is well known that the lead-shielded samples cannot be analyzed with any of the “classic” low energy analysis methods, including older versions of MGA. However, with the new high-energy mode capability that has been incorporated into the latest version of MGA described here, this limitation has been removed. 

References

R. Gunnink, “A New One-Detector Analysis Method for Rapid High-Precision Plutonium Isotopic Measurements”, Proceedings of the 9th ESARDA Symposium on Safeguards and Nuclear Material Management, London, UK, 12-14 May 1987: 167-171. 

D. D’Adamo, M. Franklin, S. Guardini, C. Vicini, R. Gunnink, W. Ruhter, and G. Varasano, "Performance Evaluation of Gamma Spectrometry Codes for International Safeguards", Proceedings of INMM, Vol XIX, 1990: 782-791.

S. Abousahl, A. Michiels, M. Bickel, R. Gunnink, and J. Verplancke, Nucl. Instr. & Meth. in Phys. Res. A368 (1996) 449-456. 

J. Verplancke, D. Davidson, M. Koskelo, R. Gunnink, J. L. Ma, J. Romeyer-Dherbey, S. Abousahl and M. Bickel, "Applying MGA for Waste Characterization". Proceedings of WM95, Feb 26 - Mar 2, 1995, Tucson, Arizona, Session 17, Paper 7. 

M. J. Koskelo and G. Gardner, “An Improved U-Pu System for Field Pu Measurements”, Proceedings of the 19^th ESARDA Annual Symposium on Safeguards and Nuclear Material Management, Montpellier, France, May 13-15, 1997: 441-445.