MGA - Multi-Group Analysis Software
|
Features
- Determines relative plutonium isotopic abundances in nondestructive assay applications
- Determines the relative amounts of other non-plutonium actinides
- Uses spectra collected with HPGe detectors which are optimized for high resolution at low energies over a wide range of count rates
- Incorporates a sophisticated peak fitting and multiplet deconvolution algorithm to improve the accuracy in samples with complex isotope mixtures
- Requires no efficiency calibration for matrix density, thickness or container characteristics
- Operates in either one or two detector mode
- Integrated into CANBERRA waste and safeguards instruments to perform a sophisticated analysis with minimal operator interaction
- Developed in collaboration with top experts in nondestructive safeguards applications
Description
The non-destructive assay of plutonium bearing samples to determine relative plutonium isotopic concentrations is one of the most common measurements in waste management and safeguards applications. However, complex samples, varying container shapes and materials and other complications have historically made it one of the more difficult measurements in the nuclear field.
Multi-Group Analysis (MGA) was designed to improve the accuracy of these measurements over traditional methods. In addition to the primary application of measuring Pu isotopics, MGA can be used to determine other actinides such as 235U, 238U, 237Np and 241Am as well.
Early public domain versions of MGA were primarily used in safeguards applications. More recently, CANBERRA has added its exclusive enhancements to improve the measurement results obtained under the difficult conditions encountered in waste applications – specifically low activities and poor counting statistics. Additional enhancements have simplified or eliminated the numerous setup and calibration steps necessary with earlier MGA versions and other methodologies. Thus it can be used in a broader range of applications involving routine, repetitive measurements taken with relatively unskilled operators.
In the single detector mode, MGA uses information from several regions of the energy spectrum that lie within an energy range of approximately 0-300 keV. In the two detector mode, MGA uses information from the low energy spectrum which has a range of 0-300 keV and from a high energy spectrum which has a range of about 0-1000 keV. The primary analysis in both cases is performed using the multiplet region at 94-104 keV. This region is a very complex multiplet, consisting of gamma ray peaks from plutonium and its progeny as well as numerous x-rays. But it is in this region where the most intense gamma ray emissions occur, thus providing the best possible detection sensitivity. MGA also uses the characteristic plutonium lines at 129 keV and 208 keV which must also be present.
To be able to unfold this complex multiplet region, MGA automatically adjusts the energy and peak shape calibration for each spectrum using peaks that are characteristic of all plutonium samples – 59 keV, 129 keV, and 208 keV. To take into account the physical processes that affect the observable gamma ray intensities at different energies, such as the detector efficiency as a function of energy, and gamma ray attenuation in absorbing materials between the sample and the detector as well as within the plutonium sample itself, MGA internally develops an intrinsic efficiency curve by evaluating 10 peaks from three isotopes.
Using the energy, shape and intrinsic efficiency information, MGA calculates a response spectrum consisting of peak energies (positions), relative peak intensities and an accurate peak shape of each peak in the 94-104 keV peak region. Accommodation is made for the Gaussian broadening of gamma ray peaks and the Lorenzian shape of the x-ray peaks. Isotopic concentrations are directly calculated for all Pu isotopes, except 242Pu which is derived from the other isotopes.
Most common applications can be accommodated with a single detector. The CANBERRA LEGe is the detector of choice in such cases due to its exceptional low energy peak shape and resolution characteristics over a wide range of count rates. Such a high performance detector is recommended for optimum performance with MGA.
In some applications, particularly those involving thick or dense container walls, the high energy spectrum collected with a typical coaxial detector is recommended to supplement the low energy MGA spectrum. In this case, in addition to the normal low energy spectrum peak regions, three high energy regions are examined and evaluated. While this results in a more complex system implementation, CANBERRA systems designed around this technique can still be operated by routine operating personnel.
Test data (summarized in Table 1) shows excellent correlation between MGA measurements and CBNM standards used in one series of tests. Accuracies shown in Table 1 are typical for safeguards samples. For waste applications, the accuracies are typically in the 20-30% range but the accuracy and whether the analysis can be performed at all is highly dependent on the waste matrix, the amount of plutonium present, and the count time used. We have shown the ability to determine the isotopic composition of a 10 mg Pu sample with a 30 minute count in a waste container.
Ordering Information
- Model S508 MGA for Windows 95/98/NT.
- Model S349 MGA for DOS and Windows 3.1.
- Model 48-0714 MGA for VMS and AXP.
| Table 1
Comparison of MGA Results with Declared Values of CBNM Standards |
||||||||||
| 238Pu | 239Pu | 240Pu | 241Pu | 241Am | ||||||
| CBNM Source |
Declared | % Diff | Declared | % Diff | Declared | % Diff | Declared | % Diff | Declared | % Diff |
| 93 | 0.0119 | –4.4 (4.0) |
93.42 | 0.03 (0.05) |
6.32 | –0.5 (0.7) |
0.223 | 0.3 (0.5) |
0.1047 | 0.4 (0.8) |
| 84 | 0.0704 | –0.6 (1.4) |
84.33 | –0.5 (0.07) |
14.21 | 0.3 (0.4) |
1.028 | –0.1 (0.5) |
0.2173 | 0.5 (0.4) |
| 70 | 0.847 | 0.5 (0.8) |
73.32 | –0.5 (0.8) |
18.30 | 1.6 (1.6) |
5.457 | –0.8 (0.5) |
1.171 | –0.8 (0.5) |
| 60 | 1.198 | 1.2 (1.2) |
62.21 | –0.5 (0.5) |
25.40 | 0.9 (1.4) |
6.68 | 1.0 (1.0) |
1.445 | 0.5 (1.0) |
| Counting time = 10 minutes, Sample = 6 g Measurement uncertainty (%) in parentheses |
||||||||||
|
| Top of Page |
Note to our International Customers: If you would like a quote on this product, please contact your Local Sales Office for assistance. |
||
| Models | Description | |
| S508C | Multi-Group Analysis Software | |
| S508CS | Multi-Group Analysis Software Support - 1 Yr | |